Up next in 10
🎥 Prompt Engineering: A Must-Have Skill for Everyone
Welcome to our essential session on Prompt Engineering – the foundational skill that unlocks the full power of Generative AI tools like ChatGPT, DALL·E, and more. Whether you're a developer, student, or professional, learning how to communicate effectively with AI is now a vital digital literacy.
🔍 Agenda:
👉 What is Prompt Engineering and why it matters in the age of Generative AI
👉 How prompts shape AI behavior, accuracy, and creativity
👉 Prompt types: Instructions, examples, context, and goals
👉 Best practices for writing effective prompts
👉 Live demos: Good vs. bad prompts – what makes the difference
👉 How Prompt Engineering varies across tools (e.g., ChatGPT, Midjourney, Copilot)
👉 Key use cases: From coding to content creation, education to productivity
👉 Challenges and limitations – What prompt engineering can’t do
👉 The future of AI interaction: Natural language as the new UI
💡 Whether you're just starting with AI or looking to sharpen your skills, this video gives you practical insights into becoming a smarter AI user.
📺 CSharp TV - Dev Streaming Destination
http://csharp.tv
Show More Show Less View Video Transcript
0:00
Feeling
0:01
Feeling
0:01
Feeling [Music]
0:15
[Music]
0:15
[Music] [Applause]
0:17
[Applause]
0:17
[Applause] [Music]
0:24
[Music]
0:24
[Music] feeling
0:26
feeling
0:26
feeling [Music]
0:28
[Music]
0:28
[Music] feeling oh
0:30
feeling oh
0:30
feeling oh [Music]
0:38
Do it.
0:42
Do it. Do it.
0:45
Do it. Do it.
0:45
Do it. Do it. [Music]
1:01
Hello. Hello. Hello. Good morning, good
1:03
Hello. Hello. Hello. Good morning, good
1:03
Hello. Hello. Hello. Good morning, good evening, good afternoon.
1:07
evening, good afternoon.
1:07
evening, good afternoon. Looks like Sarak is first one to join
1:09
Looks like Sarak is first one to join
1:09
Looks like Sarak is first one to join here. Sart, welcome to the show.
1:13
here. Sart, welcome to the show.
1:13
here. Sart, welcome to the show. My name is Mahes Chand. I'm founder of
1:14
My name is Mahes Chand. I'm founder of
1:14
My name is Mahes Chand. I'm founder of C# Corner. If you are first time joining
1:18
C# Corner. If you are first time joining
1:18
C# Corner. If you are first time joining me, welcome to the show. If you are
1:21
me, welcome to the show. If you are
1:21
me, welcome to the show. If you are joining, if it's a regular regular
1:23
joining, if it's a regular regular
1:23
joining, if it's a regular regular visitor, welcome back to the show.
1:27
visitor, welcome back to the show.
1:27
visitor, welcome back to the show. As you can see, we are going to talk
1:29
As you can see, we are going to talk
1:30
As you can see, we are going to talk about some exciting stuff here. Looks
1:31
about some exciting stuff here. Looks
1:31
about some exciting stuff here. Looks like b several people have joined us.
1:33
like b several people have joined us.
1:33
like b several people have joined us. Welcome to the show. Next, Jeie, good
1:35
Welcome to the show. Next, Jeie, good
1:35
Welcome to the show. Next, Jeie, good evening. Good evening. We have Vidwan,
1:39
evening. Good evening. We have Vidwan,
1:39
evening. Good evening. We have Vidwan, we have Atul Gupta joining us, guys.
1:42
we have Atul Gupta joining us, guys.
1:42
we have Atul Gupta joining us, guys. Welcome to the show.
1:44
Welcome to the show.
1:44
Welcome to the show. We will give it a minute and then let
1:46
We will give it a minute and then let
1:46
We will give it a minute and then let everybody join. Today is going to be
1:48
everybody join. Today is going to be
1:48
everybody join. Today is going to be very exciting session. Today
1:51
very exciting session. Today
1:51
very exciting session. Today we are going to have a lot of people
1:53
we are going to have a lot of people
1:53
we are going to have a lot of people joining us. Sizumi is joining us.
1:55
joining us. Sizumi is joining us.
1:56
joining us. Sizumi is joining us. Nishika Ravi joining us. Welcome to the
1:59
Nishika Ravi joining us. Welcome to the
1:59
Nishika Ravi joining us. Welcome to the show guys. Everybody this exciting
2:02
show guys. Everybody this exciting
2:02
show guys. Everybody this exciting stuff. Let me also try to do this. I
2:05
stuff. Let me also try to do this. I
2:05
stuff. Let me also try to do this. I don't like I know there's a virtual
2:07
don't like I know there's a virtual
2:07
don't like I know there's a virtual background.
2:09
background.
2:09
background. I will remove that because I don't like
2:11
I will remove that because I don't like
2:11
I will remove that because I don't like the my head shaking.
2:14
the my head shaking.
2:14
the my head shaking. Piluna welcome to the show. This is
2:17
Piluna welcome to the show. This is
2:17
Piluna welcome to the show. This is Mahes Chan Michael Javier. Hello,
2:19
Mahes Chan Michael Javier. Hello,
2:19
Mahes Chan Michael Javier. Hello, welcome to the show guys. Let's get
2:21
welcome to the show guys. Let's get
2:21
welcome to the show guys. Let's get started.
2:23
started.
2:23
started. Today we are going to talk about prompt
2:27
Today we are going to talk about prompt
2:27
Today we are going to talk about prompt engineering and I'm not sure if you guys
2:30
engineering and I'm not sure if you guys
2:30
engineering and I'm not sure if you guys already working with prompt engineering
2:31
already working with prompt engineering
2:31
already working with prompt engineering but I am 100% sure all of you
2:35
but I am 100% sure all of you
2:35
but I am 100% sure all of you are using chat GPT and other JAI tools.
2:41
are using chat GPT and other JAI tools.
2:41
are using chat GPT and other JAI tools. Um, in my last show, if you missed the
2:43
Um, in my last show, if you missed the
2:43
Um, in my last show, if you missed the last show, I talked about introduction
2:45
last show, I talked about introduction
2:45
last show, I talked about introduction to JNAI, generative AI. If you missed
2:48
to JNAI, generative AI. If you missed
2:48
to JNAI, generative AI. If you missed that, you can go and watch uh generative
2:51
that, you can go and watch uh generative
2:51
that, you can go and watch uh generative AI show recorded on YouTube,
2:54
AI show recorded on YouTube,
2:54
AI show recorded on YouTube, Telegram, I mean YouTube, C# Corner TV
2:57
Telegram, I mean YouTube, C# Corner TV
2:57
Telegram, I mean YouTube, C# Corner TV and bunch of other channels. Bruno is
2:59
and bunch of other channels. Bruno is
2:59
and bunch of other channels. Bruno is joining us from Canada. Bruno, welcome
3:01
joining us from Canada. Bruno, welcome
3:01
joining us from Canada. Bruno, welcome to the show. Viml Afran,
3:04
to the show. Viml Afran,
3:04
to the show. Viml Afran, we have
3:06
we have
3:06
we have Rohit Gupta joining us. Sita Sanchi
3:10
Rohit Gupta joining us. Sita Sanchi
3:10
Rohit Gupta joining us. Sita Sanchi Ambrose Sansia Ambrose welcome to the
3:13
Ambrose Sansia Ambrose welcome to the
3:13
Ambrose Sansia Ambrose welcome to the show
3:16
great this is going to be good show guys
3:18
great this is going to be good show guys
3:18
great this is going to be good show guys a lot of people have joined us from
3:20
a lot of people have joined us from
3:20
a lot of people have joined us from different places so let's talk about
3:22
different places so let's talk about
3:22
different places so let's talk about today we are going to talk about prompt
3:24
today we are going to talk about prompt
3:24
today we are going to talk about prompt engineering this is one of the skills
3:27
engineering this is one of the skills
3:27
engineering this is one of the skills everybody today everyone everyone must
3:30
everybody today everyone everyone must
3:30
everybody today everyone everyone must learn regardless of you are a computer
3:34
learn regardless of you are a computer
3:34
learn regardless of you are a computer science programmer or developer or even
3:37
science programmer or developer or even
3:37
science programmer or developer or even just a business person or trying to
3:39
just a business person or trying to
3:39
just a business person or trying to build your own thing. uh what we will
3:41
build your own thing. uh what we will
3:42
build your own thing. uh what we will talk in this show today is um
3:45
talk in this show today is um
3:45
talk in this show today is um we are going to cover uh let's look at
3:48
we are going to cover uh let's look at
3:48
we are going to cover uh let's look at the today we will talk about
3:51
the today we will talk about
3:51
the today we will talk about make the full screen here we'll talk
3:53
make the full screen here we'll talk
3:53
make the full screen here we'll talk about gener you know I think we already
3:55
about gener you know I think we already
3:55
about gener you know I think we already discussed generative AI so not too much
3:57
discussed generative AI so not too much
3:57
discussed generative AI so not too much time little little overview on jai we'll
4:00
time little little overview on jai we'll
4:00
time little little overview on jai we'll talk about a little bit LLMs and then we
4:03
talk about a little bit LLMs and then we
4:03
talk about a little bit LLMs and then we will focus on prompt engineering what is
4:05
will focus on prompt engineering what is
4:05
will focus on prompt engineering what is prompt engineering why prompt
4:07
prompt engineering why prompt
4:07
prompt engineering why prompt engineering is matters today that's the
4:09
engineering is matters today that's the
4:09
engineering is matters today that's the key point here different prompt examples
4:13
key point here different prompt examples
4:13
key point here different prompt examples we will look at what are different core
4:14
we will look at what are different core
4:14
we will look at what are different core principles and we will see some demos
4:16
principles and we will see some demos
4:16
principles and we will see some demos hands-on demos you know how to write
4:20
hands-on demos you know how to write
4:20
hands-on demos you know how to write good prompts and then we'll do Q&A so in
4:22
good prompts and then we'll do Q&A so in
4:22
good prompts and then we'll do Q&A so in the end we will do Q&A so have your
4:24
the end we will do Q&A so have your
4:24
the end we will do Q&A so have your questions ready so let's get this
4:27
questions ready so let's get this
4:27
questions ready so let's get this started
4:29
started
4:29
started but before I go there guys I want to
4:31
but before I go there guys I want to
4:31
but before I go there guys I want to tell you something that if you look at
4:33
tell you something that if you look at
4:33
tell you something that if you look at today if you are in a computer science
4:35
today if you are in a computer science
4:35
today if you are in a computer science major or you are graduating or you plan
4:39
major or you are graduating or you plan
4:39
major or you are graduating or you plan to be graduating in couple of years um
4:42
to be graduating in couple of years um
4:42
to be graduating in couple of years um your first year, second year doesn't
4:43
your first year, second year doesn't
4:43
your first year, second year doesn't matter or you are a an engineering or
4:46
matter or you are a an engineering or
4:46
matter or you are a an engineering or software developer already writing
4:48
software developer already writing
4:48
software developer already writing building applications and code prompt
4:51
building applications and code prompt
4:51
building applications and code prompt engineering is something you will need
4:53
engineering is something you will need
4:53
engineering is something you will need and this is the one of the fastest
4:56
and this is the one of the fastest
4:56
and this is the one of the fastest growing jobs in demand today.
4:59
growing jobs in demand today.
4:59
growing jobs in demand today. So this will also add to your to your
5:03
So this will also add to your to your
5:03
So this will also add to your to your skill set today after this show you will
5:05
skill set today after this show you will
5:05
skill set today after this show you will learn more about this and uh how do you
5:07
learn more about this and uh how do you
5:07
learn more about this and uh how do you get better at
5:11
brief. We already talked about jai. Jai
5:13
brief. We already talked about jai. Jai
5:13
brief. We already talked about jai. Jai is generative AI. I'm not going to spend
5:15
is generative AI. I'm not going to spend
5:15
is generative AI. I'm not going to spend much time on this today. So what is
5:17
much time on this today. So what is
5:17
much time on this today. So what is generative AI is AI? It's a part of AI
5:21
generative AI is AI? It's a part of AI
5:21
generative AI is AI? It's a part of AI is a is one of the subsets of AI that
5:25
is a is one of the subsets of AI that
5:25
is a is one of the subsets of AI that focuses on generating new content and
5:28
focuses on generating new content and
5:28
focuses on generating new content and generating AI. Generative AI usually
5:31
generating AI. Generative AI usually
5:31
generating AI. Generative AI usually text content. But then there's a um you
5:34
text content. But then there's a um you
5:34
text content. But then there's a um you can generate text, generate images,
5:37
can generate text, generate images,
5:37
can generate text, generate images, videos,
5:39
videos,
5:39
videos, um you can make songs, music, all those
5:42
um you can make songs, music, all those
5:42
um you can make songs, music, all those thing they all come under generative AI.
5:46
thing they all come under generative AI.
5:46
thing they all come under generative AI. Some of the products you're already
5:49
Some of the products you're already
5:49
Some of the products you're already using, Chat, GPT, Copilot, MidJourney,
5:51
using, Chat, GPT, Copilot, MidJourney,
5:51
using, Chat, GPT, Copilot, MidJourney, Sora, they all come under generative AI.
5:55
Sora, they all come under generative AI.
5:56
Sora, they all come under generative AI. In generative AI, the brain behind
5:59
In generative AI, the brain behind
5:59
In generative AI, the brain behind generative AI is this something called
6:01
generative AI is this something called
6:01
generative AI is this something called LLM large language model. And there are
6:04
LLM large language model. And there are
6:04
LLM large language model. And there are bunch of large language models. Okay?
6:08
bunch of large language models. Okay?
6:08
bunch of large language models. Okay? And we talk this little bit but by next
6:10
And we talk this little bit but by next
6:10
And we talk this little bit but by next episode in my coming episodes I will
6:13
episode in my coming episodes I will
6:13
episode in my coming episodes I will focus one episode just on LLM. I will
6:17
focus one episode just on LLM. I will
6:17
focus one episode just on LLM. I will talk about what is LM, what's behind the
6:19
talk about what is LM, what's behind the
6:19
talk about what is LM, what's behind the scene, how the LLM look like when you
6:21
scene, how the LLM look like when you
6:21
scene, how the LLM look like when you look at the from the technology and
6:23
look at the from the technology and
6:23
look at the from the technology and architectural perspective. how to even
6:26
architectural perspective. how to even
6:26
architectural perspective. how to even build your LLM. We will discuss all that
6:28
build your LLM. We will discuss all that
6:28
build your LLM. We will discuss all that in coming time. Today's focus is prompt
6:30
in coming time. Today's focus is prompt
6:30
in coming time. Today's focus is prompt engineering.
6:32
engineering.
6:32
engineering. So when prompt engineer you think of
6:35
So when prompt engineer you think of
6:35
So when prompt engineer you think of there are three parts of this AI we are
6:37
there are three parts of this AI we are
6:37
there are three parts of this AI we are talking about generative AI. One is
6:39
talking about generative AI. One is
6:39
talking about generative AI. One is input and there's something called LLM
6:42
input and there's something called LLM
6:42
input and there's something called LLM and then output. So when you go to chat
6:45
and then output. So when you go to chat
6:45
and then output. So when you go to chat GPT and say hey create an email or you
6:49
GPT and say hey create an email or you
6:49
GPT and say hey create an email or you know you go to midjourney or copilot and
6:52
know you go to midjourney or copilot and
6:52
know you go to midjourney or copilot and say create an image or create a logo
6:54
say create an image or create a logo
6:54
say create an image or create a logo that's your input then genai does llms
6:58
that's your input then genai does llms
6:58
that's your input then genai does llms and that does some stuff and then
7:01
and that does some stuff and then
7:01
and that does some stuff and then generates some output and output is your
7:04
generates some output and output is your
7:04
generates some output and output is your you know what you asked for.
7:07
you know what you asked for.
7:07
you know what you asked for. So
7:09
So
7:09
So the all magic happens in this this
7:13
the all magic happens in this this
7:14
the all magic happens in this this product. It's it's a think of like a
7:15
product. It's it's a think of like a
7:15
product. It's it's a think of like a computer program programs bunch of
7:18
computer program programs bunch of
7:18
computer program programs bunch of programs which I'll explain you more.
7:20
programs which I'll explain you more.
7:20
programs which I'll explain you more. They do they look at your prompt they do
7:22
They do they look at your prompt they do
7:22
They do they look at your prompt they do the processing and all that and they
7:24
the processing and all that and they
7:24
the processing and all that and they creates the content which is text which
7:26
creates the content which is text which
7:26
creates the content which is text which is could be you can just say for example
7:29
is could be you can just say for example
7:29
is could be you can just say for example you know draw a cat when you say draw a
7:31
you know draw a cat when you say draw a
7:31
you know draw a cat when you say draw a cat it's input llm take that understand
7:33
cat it's input llm take that understand
7:33
cat it's input llm take that understand what you're trying to do it creates the
7:35
what you're trying to do it creates the
7:35
what you're trying to do it creates the image and spits out the image that looks
7:38
image and spits out the image that looks
7:38
image and spits out the image that looks like a cat right that makes sense so LLM
7:40
like a cat right that makes sense so LLM
7:40
like a cat right that makes sense so LLM is the thing here it's called large
7:42
is the thing here it's called large
7:42
is the thing here it's called large language model and this is the brain
7:44
language model and this is the brain
7:44
language model and this is the brain behind all jai tools such as chad GPT
7:49
behind all jai tools such as chad GPT
7:49
behind all jai tools such as chad GPT so they're going to be a following show
7:50
so they're going to be a following show
7:50
so they're going to be a following show just on a LLM and that's going to be a
7:53
just on a LLM and that's going to be a
7:53
just on a LLM and that's going to be a big one because this is a little complex
7:55
big one because this is a little complex
7:55
big one because this is a little complex topics understanding LLMs I'm sure some
7:58
topics understanding LLMs I'm sure some
7:58
topics understanding LLMs I'm sure some of you have already read and learn about
8:00
of you have already read and learn about
8:00
of you have already read and learn about this
8:02
this
8:02
this so let's come to the uh topic today
8:04
so let's come to the uh topic today
8:04
so let's come to the uh topic today today's topic is prompt engineering so
8:06
today's topic is prompt engineering so
8:06
today's topic is prompt engineering so what is prompt engineering
8:09
what is prompt engineering
8:09
what is prompt engineering in simple terms prompt engineering is a
8:12
in simple terms prompt engineering is a
8:12
in simple terms prompt engineering is a is a I this craft or this art or this
8:15
is a I this craft or this art or this
8:15
is a I this craft or this art or this the way you create these prompts the way
8:18
the way you create these prompts the way
8:18
the way you create these prompts the way you ask a question
8:21
you ask a question
8:21
you ask a question to a tool like chat GPT that is called
8:24
to a tool like chat GPT that is called
8:24
to a tool like chat GPT that is called prompt engineering. Prompt engineering
8:26
prompt engineering. Prompt engineering
8:26
prompt engineering. Prompt engineering not really engineering. It's not really
8:28
not really engineering. It's not really
8:28
not really engineering. It's not really something where you have to do with the
8:30
something where you have to do with the
8:30
something where you have to do with the hands and you have to draw build
8:32
hands and you have to draw build
8:32
hands and you have to draw build something. It's really building the
8:33
something. It's really building the
8:34
something. It's really building the prompt. Prompt engineering is creating
8:36
prompt. Prompt engineering is creating
8:36
prompt. Prompt engineering is creating the prompt, building the prompt, right?
8:39
the prompt, building the prompt, right?
8:39
the prompt, building the prompt, right? So the input, the input ji needs, how
8:42
So the input, the input ji needs, how
8:42
So the input, the input ji needs, how you craft that, that is prompt
8:44
you craft that, that is prompt
8:44
you craft that, that is prompt engineering and learning the skill of
8:46
engineering and learning the skill of
8:46
engineering and learning the skill of that creating the crafting that that
8:48
that creating the crafting that that
8:48
that creating the crafting that that prompt is the key to get most out of
8:52
prompt is the key to get most out of
8:52
prompt is the key to get most out of these AI tools. So it's literally a art
8:57
these AI tools. So it's literally a art
8:57
these AI tools. So it's literally a art the way you craft that. So in other
9:00
the way you craft that. So in other
9:00
the way you craft that. So in other words, prompt engineering like asking
9:04
words, prompt engineering like asking
9:04
words, prompt engineering like asking AI a smart question. Smarter the
9:06
AI a smart question. Smarter the
9:06
AI a smart question. Smarter the question, better output you will get. So
9:09
question, better output you will get. So
9:09
question, better output you will get. So like you have to become smart to even
9:11
like you have to become smart to even
9:11
like you have to become smart to even ask the question. That's what really
9:12
ask the question. That's what really
9:12
ask the question. That's what really prompt engineering is all about. You can
9:14
prompt engineering is all about. You can
9:14
prompt engineering is all about. You can ask general question, right? You can
9:15
ask general question, right? You can
9:15
ask general question, right? You can just go and say hey generate image of
9:18
just go and say hey generate image of
9:18
just go and say hey generate image of cat. Yeah, I will go and generate a cat.
9:20
cat. Yeah, I will go and generate a cat.
9:20
cat. Yeah, I will go and generate a cat. But what if you want a black cat? What
9:22
But what if you want a black cat? What
9:22
But what if you want a black cat? What if you want a specific kind of cat? What
9:25
if you want a specific kind of cat? What
9:25
if you want a specific kind of cat? What if you want a three-year-old cat? So all
9:28
if you want a three-year-old cat? So all
9:28
if you want a three-year-old cat? So all those combined what if your cat is
9:29
those combined what if your cat is
9:29
those combined what if your cat is wearing a hat something. So learning the
9:32
wearing a hat something. So learning the
9:32
wearing a hat something. So learning the craft of this whole prompting or this
9:36
craft of this whole prompting or this
9:36
craft of this whole prompting or this creating this input that your AI would
9:38
creating this input that your AI would
9:38
creating this input that your AI would understand that the AI would understand
9:41
understand that the AI would understand
9:41
understand that the AI would understand that is prompt engineering.
9:43
that is prompt engineering.
9:43
that is prompt engineering. Guys keep asking questions. get your
9:45
Guys keep asking questions. get your
9:45
Guys keep asking questions. get your questions ready because I what I do is
9:48
questions ready because I what I do is
9:48
questions ready because I what I do is in these shows if you're joining me
9:50
in these shows if you're joining me
9:50
in these shows if you're joining me first time I go through these concepts
9:52
first time I go through these concepts
9:52
first time I go through these concepts and these ideas and demo some things but
9:54
and these ideas and demo some things but
9:54
and these ideas and demo some things but then I more focus on questions and I
9:56
then I more focus on questions and I
9:56
then I more focus on questions and I want to make sure I ask I answer most of
10:00
want to make sure I ask I answer most of
10:00
want to make sure I ask I answer most of your questions questions can be about
10:01
your questions questions can be about
10:01
your questions questions can be about this topic prompt engineering question
10:03
this topic prompt engineering question
10:03
this topic prompt engineering question can be about the jobs question can be
10:05
can be about the jobs question can be
10:05
can be about the jobs question can be about career or anything you have
10:07
about career or anything you have
10:07
about career or anything you have anything about AI today we will focus on
10:10
anything about AI today we will focus on
10:10
anything about AI today we will focus on prompt engineering so let those
10:11
prompt engineering so let those
10:11
prompt engineering so let those questions coming in the end of this
10:14
questions coming in the end of this
10:14
questions coming in the end of this after 10 15 minutes or so we will just
10:16
after 10 15 minutes or so we will just
10:16
after 10 15 minutes or so we will just take your questions.
10:20
take your questions.
10:20
take your questions. So hopefully now you understand what
10:21
So hopefully now you understand what
10:21
So hopefully now you understand what prompt engineering is. Pretty simple.
10:24
prompt engineering is. Pretty simple.
10:24
prompt engineering is. Pretty simple. The art of crafting the input that Jai
10:27
The art of crafting the input that Jai
10:27
The art of crafting the input that Jai needs to generate perfect output for
10:30
needs to generate perfect output for
10:30
needs to generate perfect output for you. That is called prompt engineering.
10:33
you. That is called prompt engineering.
10:33
you. That is called prompt engineering. And you can decide def the way you want
10:35
And you can decide def the way you want
10:35
And you can decide def the way you want your own definition. I create my own
10:36
your own definition. I create my own
10:36
your own definition. I create my own definitions all the time.
10:39
definitions all the time.
10:39
definitions all the time. So
10:41
So
10:41
So this this show is brought to you by SH
10:43
this this show is brought to you by SH
10:43
this this show is brought to you by SH token. If you guys are not familiar with
10:45
token. If you guys are not familiar with
10:45
token. If you guys are not familiar with SH token I really want you to check out
10:48
SH token I really want you to check out
10:48
SH token I really want you to check out this check out a website called shar
10:51
this check out a website called shar
10:51
this check out a website called shar economy.org. It is a reward token for
10:54
economy.org. It is a reward token for
10:54
economy.org. It is a reward token for communities like C# corner. Many of you
10:57
communities like C# corner. Many of you
10:57
communities like C# corner. Many of you are member of Corner. So what does that
10:59
are member of Corner. So what does that
10:59
are member of Corner. So what does that mean? Being a member of C# Corner if
11:02
mean? Being a member of C# Corner if
11:02
mean? Being a member of C# Corner if anything you do on the site you say you
11:05
anything you do on the site you say you
11:05
anything you do on the site you say you are learning something you are going to
11:06
are learning something you are going to
11:06
are learning something you are going to earn tokens. you are um sharing
11:11
earn tokens. you are um sharing
11:11
earn tokens. you are um sharing something, you are posting something,
11:12
something, you are posting something,
11:12
something, you are posting something, you are speaking, you are even watching
11:14
you are speaking, you are even watching
11:14
you are speaking, you are even watching this show today. I eventually you're
11:16
this show today. I eventually you're
11:16
this show today. I eventually you're going to start earning tokens for any
11:18
going to start earning tokens for any
11:18
going to start earning tokens for any involvement with the with the community.
11:21
involvement with the with the community.
11:21
involvement with the with the community. You're asking question, you're replying,
11:23
You're asking question, you're replying,
11:23
You're asking question, you're replying, you're posting question, you're sharing
11:24
you're posting question, you're sharing
11:24
you're posting question, you're sharing your code on C# Corner. So check out C#c
11:27
your code on C# Corner. So check out C#c
11:27
your code on C# Corner. So check out C#c corner.com
11:29
corner.com
11:29
corner.com and we also have bounties on there. Now
11:31
and we also have bounties on there. Now
11:31
and we also have bounties on there. Now you can earn bounties.
11:33
you can earn bounties.
11:33
you can earn bounties. Well, it's free like you just start
11:35
Well, it's free like you just start
11:35
Well, it's free like you just start participating. There's a there bounty
11:37
participating. There's a there bounty
11:37
participating. There's a there bounty where you can win cash prize, you can
11:39
where you can win cash prize, you can
11:39
where you can win cash prize, you can win tokens. Um there is a app called
11:41
win tokens. Um there is a app called
11:41
win tokens. Um there is a app called Sharp Rewards app. You go to this app
11:44
Sharp Rewards app. You go to this app
11:44
Sharp Rewards app. You go to this app store, Google or Apple App Store and
11:46
store, Google or Apple App Store and
11:46
store, Google or Apple App Store and search for Sharp Rewards. Download this
11:48
search for Sharp Rewards. Download this
11:48
search for Sharp Rewards. Download this app. In this app, you can earn these
11:51
app. In this app, you can earn these
11:51
app. In this app, you can earn these tokens free every day by doing some
11:53
tokens free every day by doing some
11:53
tokens free every day by doing some staking skills challenges by being
11:56
staking skills challenges by being
11:56
staking skills challenges by being healthy, walking. There are different
11:58
healthy, walking. There are different
11:58
healthy, walking. There are different ways to earn those tokens. Check it out.
12:01
ways to earn those tokens. Check it out.
12:01
ways to earn those tokens. Check it out. And if you are in a crypto guys and you
12:03
And if you are in a crypto guys and you
12:03
And if you are in a crypto guys and you say, you know what, I like Sharp token.
12:05
say, you know what, I like Sharp token.
12:05
say, you know what, I like Sharp token. I want to buy it. It's available on two
12:08
I want to buy it. It's available on two
12:08
I want to buy it. It's available on two exchange coin store and bitmark. So
12:11
exchange coin store and bitmark. So
12:11
exchange coin store and bitmark. So check it out. Any questions feel free to
12:13
check it out. Any questions feel free to
12:13
check it out. Any questions feel free to reach out to us on C# phone as well.
12:17
reach out to us on C# phone as well.
12:17
reach out to us on C# phone as well. All right. So let's back to this prompt
12:20
All right. So let's back to this prompt
12:20
All right. So let's back to this prompt engineering. So I hope you know what
12:22
engineering. So I hope you know what
12:22
engineering. So I hope you know what prompt engineering by now. If you still
12:24
prompt engineering by now. If you still
12:24
prompt engineering by now. If you still have confusion just ask your questions.
12:27
have confusion just ask your questions.
12:27
have confusion just ask your questions. And I'm not watching your questions
12:29
And I'm not watching your questions
12:29
And I'm not watching your questions right now because I want to first finish
12:31
right now because I want to first finish
12:31
right now because I want to first finish my talk and then I will go through these
12:33
my talk and then I will go through these
12:33
my talk and then I will go through these questions.
12:34
questions.
12:34
questions. So why prompt engineer matters today
12:37
So why prompt engineer matters today
12:37
So why prompt engineer matters today like right now?
12:39
like right now?
12:40
like right now? So the important part here is the reason
12:42
So the important part here is the reason
12:42
So the important part here is the reason is prompt engineering matter is that the
12:44
is prompt engineering matter is that the
12:44
is prompt engineering matter is that the because and I'll give you an example
12:46
because and I'll give you an example
12:46
because and I'll give you an example right. So if you go to even at your home
12:49
right. So if you go to even at your home
12:49
right. So if you go to even at your home right your home and you have a cook you
12:53
right your home and you have a cook you
12:53
right your home and you have a cook you know a lot of people have cook at home I
12:55
know a lot of people have cook at home I
12:55
know a lot of people have cook at home I think every time I go there's several
12:56
think every time I go there's several
12:56
think every time I go there's several cooks at home morning cook is different
12:59
cooks at home morning cook is different
12:59
cooks at home morning cook is different evening is different right? So if you
13:01
evening is different right? So if you
13:01
evening is different right? So if you tell a cook make me food
13:05
tell a cook make me food
13:05
tell a cook make me food just simple say make me food what does
13:07
just simple say make me food what does
13:07
just simple say make me food what does that mean she doesn't know what to cook
13:10
that mean she doesn't know what to cook
13:10
that mean she doesn't know what to cook when she'll make whatever food and
13:12
when she'll make whatever food and
13:12
when she'll make whatever food and they'll say oh I don't like this food
13:14
they'll say oh I don't like this food
13:14
they'll say oh I don't like this food can you make again
13:16
can you make again
13:16
can you make again so that's one instruction you are giving
13:18
so that's one instruction you are giving
13:18
so that's one instruction you are giving to your cook is make you food second is
13:21
to your cook is make you food second is
13:21
to your cook is make you food second is can you cook
13:23
can you cook
13:23
can you cook can you put u cauliflower potato and
13:26
can you put u cauliflower potato and
13:26
can you put u cauliflower potato and this much and put this you know this
13:28
this much and put this you know this
13:28
this much and put this you know this much spice make it medium spice
13:30
much spice make it medium spice
13:30
much spice make it medium spice sometimes When you go to a high-end
13:32
sometimes When you go to a high-end
13:32
sometimes When you go to a high-end restaurant, when you go to a high-end
13:34
restaurant, when you go to a high-end
13:34
restaurant, when you go to a high-end restaurant, they will ask you, "Oh, you
13:36
restaurant, they will ask you, "Oh, you
13:36
restaurant, they will ask you, "Oh, you want what do you want exactly? How you
13:38
want what do you want exactly? How you
13:38
want what do you want exactly? How you want medium spicy, low spicy?" All those
13:41
want medium spicy, low spicy?" All those
13:41
want medium spicy, low spicy?" All those things. More details you tell, better
13:44
things. More details you tell, better
13:44
things. More details you tell, better food you will get, right? That's what it
13:47
food you will get, right? That's what it
13:47
food you will get, right? That's what it is. Simple. Think about a web developer.
13:49
is. Simple. Think about a web developer.
13:49
is. Simple. Think about a web developer. If you're a web developer or you are
13:51
If you're a web developer or you are
13:51
If you're a web developer or you are designing a website and client comes to
13:53
designing a website and client comes to
13:53
designing a website and client comes to you and says, "Create me a website."
13:55
you and says, "Create me a website."
13:56
you and says, "Create me a website." What does that mean really? That means
13:57
What does that mean really? That means
13:57
What does that mean really? That means nothing. It's like you just create
13:59
nothing. It's like you just create
13:59
nothing. It's like you just create whatever website. Okay? You create a web
14:01
whatever website. Okay? You create a web
14:01
whatever website. Okay? You create a web page with whatever but client will say
14:04
page with whatever but client will say
14:04
page with whatever but client will say create me a website that will have a
14:05
create me a website that will have a
14:06
create me a website that will have a user registration. People can there's a
14:07
user registration. People can there's a
14:08
user registration. People can there's a landing page where they will learn about
14:09
landing page where they will learn about
14:09
landing page where they will learn about my product. Product is all about selling
14:12
my product. Product is all about selling
14:12
my product. Product is all about selling cups and you should also put two four
14:15
cups and you should also put two four
14:15
cups and you should also put two four pictures of my cups. These are the
14:18
pictures of my cups. These are the
14:18
pictures of my cups. These are the pictures. Can you upload them? And then
14:20
pictures. Can you upload them? And then
14:20
pictures. Can you upload them? And then talk about details what the pricing is.
14:22
talk about details what the pricing is.
14:22
talk about details what the pricing is. So more details you tell to a developer
14:26
So more details you tell to a developer
14:26
So more details you tell to a developer they will come prepared and build the
14:28
they will come prepared and build the
14:28
they will come prepared and build the website or design the website the way
14:30
website or design the website the way
14:30
website or design the website the way you want it. So that's kind of
14:33
you want it. So that's kind of
14:33
you want it. So that's kind of difference here. Do you see the
14:34
difference here. Do you see the
14:34
difference here. Do you see the difference?
14:36
difference?
14:36
difference? The simple way requirements detail the
14:40
The simple way requirements detail the
14:40
The simple way requirements detail the requirements are better software you
14:42
requirements are better software you
14:42
requirements are better software you will get developed by the team because
14:44
will get developed by the team because
14:44
will get developed by the team because now they are more clear on the
14:45
now they are more clear on the
14:45
now they are more clear on the requirement. This is the same exact way
14:48
requirement. This is the same exact way
14:48
requirement. This is the same exact way AI works. when you're talking to all
14:51
AI works. when you're talking to all
14:51
AI works. when you're talking to all these geni tools like chat GPT or
14:53
these geni tools like chat GPT or
14:53
these geni tools like chat GPT or copilot
14:55
copilot
14:55
copilot better requirements you have which is
14:57
better requirements you have which is
14:57
better requirements you have which is called prompt in this case better output
15:00
called prompt in this case better output
15:00
called prompt in this case better output desired output you will get so that's
15:03
desired output you will get so that's
15:03
desired output you will get so that's exactly what it is well it's that's very
15:05
exactly what it is well it's that's very
15:05
exactly what it is well it's that's very important to learn prompt engineering
15:07
important to learn prompt engineering
15:07
important to learn prompt engineering because better you are at that more
15:10
because better you are at that more
15:10
because better you are at that more details you provide more understand how
15:14
details you provide more understand how
15:14
details you provide more understand how geni is thinking better you can ask the
15:17
geni is thinking better you can ask the
15:17
geni is thinking better you can ask the question craft the question better and
15:19
question craft the question better and
15:19
question craft the question better and faster results you will get.
15:22
faster results you will get.
15:22
faster results you will get. That make sense?
15:25
That make sense?
15:25
That make sense? So simple words look at these things
15:27
So simple words look at these things
15:27
So simple words look at these things like if you AI models the thing with AI
15:30
like if you AI models the thing with AI
15:30
like if you AI models the thing with AI models is that they're not smart. They
15:33
models is that they're not smart. They
15:33
models is that they're not smart. They are dumb. Okay, they don't think what
15:34
are dumb. Okay, they don't think what
15:34
are dumb. Okay, they don't think what you are really want. They don't know
15:37
you are really want. They don't know
15:37
you are really want. They don't know because they can see you. They can talk
15:39
because they can see you. They can talk
15:39
because they can see you. They can talk to you. They don't see your expressions.
15:41
to you. They don't see your expressions.
15:41
to you. They don't see your expressions. So what they only know what you're
15:43
So what they only know what you're
15:43
So what they only know what you're typing there. So more details you
15:46
typing there. So more details you
15:46
typing there. So more details you provide in your input prompt better
15:49
provide in your input prompt better
15:49
provide in your input prompt better results they will get you. Okay. So
15:52
results they will get you. Okay. So
15:52
results they will get you. Okay. So these are some of the benefits right. So
15:54
these are some of the benefits right. So
15:54
these are some of the benefits right. So better results from same AI like I can
15:57
better results from same AI like I can
15:57
better results from same AI like I can ask same question and you can ask same
15:59
ask same question and you can ask same
15:59
ask same question and you can ask same question in different ways is the same
16:01
question in different ways is the same
16:01
question in different ways is the same AI but because your prompt was better
16:04
AI but because your prompt was better
16:04
AI but because your prompt was better your results maybe will be better. It
16:07
your results maybe will be better. It
16:07
your results maybe will be better. It saves time right. It saves time. If you
16:09
saves time right. It saves time. If you
16:09
saves time right. It saves time. If you already know everything in details, it
16:11
already know everything in details, it
16:11
already know everything in details, it will save everybody time and there's
16:13
will save everybody time and there's
16:13
will save everybody time and there's less work behind the scene with AI. Why?
16:16
less work behind the scene with AI. Why?
16:16
less work behind the scene with AI. Why? Because if you don't get desired output,
16:18
Because if you don't get desired output,
16:18
Because if you don't get desired output, you're going to ask again and then AI
16:20
you're going to ask again and then AI
16:20
you're going to ask again and then AI have to do work again.
16:22
have to do work again.
16:22
have to do work again. It you can also control the format, the
16:25
It you can also control the format, the
16:25
It you can also control the format, the kind of output format you want, what
16:27
kind of output format you want, what
16:27
kind of output format you want, what style you want, what tone you want. You
16:29
style you want, what tone you want. You
16:29
style you want, what tone you want. You can even tell AI that oh write me a poem
16:33
can even tell AI that oh write me a poem
16:33
can even tell AI that oh write me a poem or write me a joke or write an email
16:35
or write me a joke or write an email
16:35
or write me a joke or write an email which is funny or which is friendly or
16:38
which is funny or which is friendly or
16:38
which is funny or which is friendly or which is professional. You can have
16:39
which is professional. You can have
16:39
which is professional. You can have these different tones right you can have
16:41
these different tones right you can have
16:41
these different tones right you can have all those details in this we will talk
16:43
all those details in this we will talk
16:43
all those details in this we will talk in my following you know talks we'll
16:45
in my following you know talks we'll
16:45
in my following you know talks we'll cover a little bit here today that's
16:47
cover a little bit here today that's
16:47
cover a little bit here today that's what really prompt engineering means
16:49
what really prompt engineering means
16:49
what really prompt engineering means understanding how LLMs work and what
16:52
understanding how LLMs work and what
16:52
understanding how LLMs work and what they expect and how to get them desired
16:54
they expect and how to get them desired
16:54
they expect and how to get them desired output from them
16:56
output from them
16:56
output from them and then you can also avoid mistakes
16:58
and then you can also avoid mistakes
16:58
and then you can also avoid mistakes because some LLM sometime they can make
17:00
because some LLM sometime they can make
17:00
because some LLM sometime they can make mistakes too so by giving them these
17:02
mistakes too so by giving them these
17:02
mistakes too so by giving them these good instructions the prompt you can
17:05
good instructions the prompt you can
17:05
good instructions the prompt you can also avoid these mistakes
17:07
also avoid these mistakes
17:07
also avoid these mistakes in less costly. So keep in mind when you
17:10
in less costly. So keep in mind when you
17:10
in less costly. So keep in mind when you ask anything to do with LLM there's a
17:13
ask anything to do with LLM there's a
17:13
ask anything to do with LLM there's a cost involved not only time not only
17:16
cost involved not only time not only
17:16
cost involved not only time not only your time not only machines time but
17:18
your time not only machines time but
17:18
your time not only machines time but every time you say something the input
17:21
every time you say something the input
17:21
every time you say something the input goes to these machines LLMs for example
17:23
goes to these machines LLMs for example
17:23
goes to these machines LLMs for example OPI opai GPT4 it has trillions of
17:27
OPI opai GPT4 it has trillions of
17:27
OPI opai GPT4 it has trillions of parameters now it's processing a lot of
17:29
parameters now it's processing a lot of
17:29
parameters now it's processing a lot of things there was a research came out
17:31
things there was a research came out
17:31
things there was a research came out that by just saying hello in the
17:33
that by just saying hello in the
17:33
that by just saying hello in the beginning of your sentence
17:35
beginning of your sentence
17:35
beginning of your sentence uh of asking a questions to chat GPT
17:38
uh of asking a questions to chat GPT
17:38
uh of asking a questions to chat GPT cost them millions of dollars each
17:40
cost them millions of dollars each
17:40
cost them millions of dollars each month. Can you imagine that? So this is
17:43
month. Can you imagine that? So this is
17:43
month. Can you imagine that? So this is exactly what I'm talking about is
17:45
exactly what I'm talking about is
17:45
exactly what I'm talking about is crafting the prompt and that's where we
17:47
crafting the prompt and that's where we
17:47
crafting the prompt and that's where we are going to be really get into more
17:49
are going to be really get into more
17:49
are going to be really get into more details in this uh this episode and
17:51
details in this uh this episode and
17:51
details in this uh this episode and upcoming following episode as well.
17:55
upcoming following episode as well.
17:55
upcoming following episode as well. I hope you're learning something. I hope
17:57
I hope you're learning something. I hope
17:57
I hope you're learning something. I hope I'm clarifying and explaining this very
17:59
I'm clarifying and explaining this very
17:59
I'm clarifying and explaining this very simple way. And if you are still not
18:01
simple way. And if you are still not
18:01
simple way. And if you are still not confused and don't know, just ask
18:03
confused and don't know, just ask
18:03
confused and don't know, just ask questions. Okay? I will take your
18:05
questions. Okay? I will take your
18:05
questions. Okay? I will take your questions in that.
18:07
questions in that.
18:07
questions in that. So let's take example these are the two
18:09
So let's take example these are the two
18:09
So let's take example these are the two examples of prompt here so you can
18:12
examples of prompt here so you can
18:12
examples of prompt here so you can understand better let's you know say I
18:14
understand better let's you know say I
18:14
understand better let's you know say I want to write a blog about electric cars
18:17
want to write a blog about electric cars
18:17
want to write a blog about electric cars right say I want to write a blog about
18:18
right say I want to write a blog about
18:18
right say I want to write a blog about electric cars I go to chat GPT and say
18:21
electric cars I go to chat GPT and say
18:21
electric cars I go to chat GPT and say write about electric cars that's one
18:23
write about electric cars that's one
18:23
write about electric cars that's one prompt
18:25
prompt
18:25
prompt now it will it will write whatever and
18:27
now it will it will write whatever and
18:27
now it will it will write whatever and we will see some examples in the end
18:29
we will see some examples in the end
18:29
we will see some examples in the end when I do demo right but in the second
18:32
when I do demo right but in the second
18:32
when I do demo right but in the second part you can say oh write a three
18:35
part you can say oh write a three
18:35
part you can say oh write a three paragraph blog introductions about the
18:37
paragraph blog introductions about the
18:37
paragraph blog introductions about the benefit of cars for city drivers
18:40
benefit of cars for city drivers
18:40
benefit of cars for city drivers including point of cost saving blah blah
18:42
including point of cost saving blah blah
18:42
including point of cost saving blah blah blah. Oh, take an example as a this you
18:45
blah. Oh, take an example as a this you
18:45
blah. Oh, take an example as a this you know Tesla more you give now AI is
18:50
know Tesla more you give now AI is
18:50
know Tesla more you give now AI is thinking and giving you the desired
18:52
thinking and giving you the desired
18:52
thinking and giving you the desired output.
18:54
output.
18:54
output. So that's exactly what prompt
18:56
So that's exactly what prompt
18:56
So that's exactly what prompt engineering is all about. Second example
18:58
engineering is all about. Second example
18:58
engineering is all about. Second example let's say write an email. You want to
19:00
let's say write an email. You want to
19:00
let's say write an email. You want to write an email to a customer about who's
19:03
write an email to a customer about who's
19:03
write an email to a customer about who's asking about pricing.
19:05
asking about pricing.
19:05
asking about pricing. So one thing is you can have a general
19:07
So one thing is you can have a general
19:07
So one thing is you can have a general saying write an email about pricing
19:09
saying write an email about pricing
19:09
saying write an email about pricing that's so it will write just generic
19:11
that's so it will write just generic
19:11
that's so it will write just generic answer but you can be specific write a
19:14
answer but you can be specific write a
19:14
answer but you can be specific write a friendly follow-up email to a customer
19:16
friendly follow-up email to a customer
19:16
friendly follow-up email to a customer who ask about our car pricing this is
19:19
who ask about our car pricing this is
19:19
who ask about our car pricing this is our website these are the pricing there
19:21
our website these are the pricing there
19:21
our website these are the pricing there are three different models and they ask
19:24
are three different models and they ask
19:24
are three different models and they ask them schedule a call this is the URL now
19:29
them schedule a call this is the URL now
19:29
them schedule a call this is the URL now chat GP is going to craft this complete
19:31
chat GP is going to craft this complete
19:31
chat GP is going to craft this complete email which will have all these details
19:34
email which will have all these details
19:34
email which will have all these details so the the Second prompt is the prompt
19:36
so the the Second prompt is the prompt
19:36
so the the Second prompt is the prompt you want to write.
19:40
Okay. So hopefully you learned what is
19:42
Okay. So hopefully you learned what is
19:42
Okay. So hopefully you learned what is prompt engineering and why it's
19:43
prompt engineering and why it's
19:44
prompt engineering and why it's important and what is difference between
19:45
important and what is difference between
19:45
important and what is difference between a weak prompt and a strong prompt. Right
19:48
a weak prompt and a strong prompt. Right
19:48
a weak prompt and a strong prompt. Right now let's think about core principles.
19:51
now let's think about core principles.
19:51
now let's think about core principles. There are some core principles when
19:53
There are some core principles when
19:53
There are some core principles when you're writing good prompt and these are
19:55
you're writing good prompt and these are
19:55
you're writing good prompt and these are simple terms and I'm going to do more
19:57
simple terms and I'm going to do more
19:57
simple terms and I'm going to do more details in my following show but this is
19:59
details in my following show but this is
19:59
details in my following show but this is very simple everybody can understand. So
20:02
very simple everybody can understand. So
20:02
very simple everybody can understand. So one thing you want to be with AI is that
20:05
one thing you want to be with AI is that
20:05
one thing you want to be with AI is that you want to be very specific but direct.
20:08
you want to be very specific but direct.
20:08
you want to be very specific but direct. Don't be too formal. Don't write that
20:10
Don't be too formal. Don't write that
20:10
Don't be too formal. Don't write that good morning blah blah blah. Just be
20:12
good morning blah blah blah. Just be
20:12
good morning blah blah blah. Just be straightforward. And if one example is
20:15
straightforward. And if one example is
20:15
straightforward. And if one example is here
20:17
here
20:17
here help me with authentication
20:19
help me with authentication
20:19
help me with authentication that's good. It's direct but it's not
20:22
that's good. It's direct but it's not
20:22
that's good. It's direct but it's not very specific. So authentic
20:25
very specific. So authentic
20:25
very specific. So authentic authentication can be multiple kinds
20:27
authentication can be multiple kinds
20:27
authentication can be multiple kinds different ways. What does that mean? Um
20:30
different ways. What does that mean? Um
20:30
different ways. What does that mean? Um or you can just be specific. You are
20:32
or you can just be specific. You are
20:32
or you can just be specific. You are programmer saying create a JW to
20:34
programmer saying create a JW to
20:34
programmer saying create a JW to middleware function for express JS that
20:36
middleware function for express JS that
20:36
middleware function for express JS that validates tokens extract user ID blah
20:39
validates tokens extract user ID blah
20:39
validates tokens extract user ID blah blah blah blah blah all those things. So
20:42
blah blah blah blah all those things. So
20:42
blah blah blah blah all those things. So that's simple you are being specific and
20:45
that's simple you are being specific and
20:46
that's simple you are being specific and also direct.
20:48
also direct.
20:48
also direct. So now the output is you are going to
20:50
So now the output is you are going to
20:50
So now the output is you are going to get is from the a from the AI is very
20:54
get is from the a from the AI is very
20:54
get is from the a from the AI is very specific as well. So you need to think
20:56
specific as well. So you need to think
20:56
specific as well. So you need to think AI as your coworker. Somebody who knows
21:01
AI as your coworker. Somebody who knows
21:01
AI as your coworker. Somebody who knows a lot, who can do a lot of work for you,
21:03
a lot, who can do a lot of work for you,
21:03
a lot, who can do a lot of work for you, but they are not smart enough to think
21:05
but they are not smart enough to think
21:05
but they are not smart enough to think about themselves like what exactly you
21:07
about themselves like what exactly you
21:07
about themselves like what exactly you want. But they will do. But if you give
21:10
want. But they will do. But if you give
21:10
want. But they will do. But if you give them more detailed instructions, they
21:12
them more detailed instructions, they
21:12
them more detailed instructions, they will do a lot for you. They have this
21:14
will do a lot for you. They have this
21:14
will do a lot for you. They have this unlimited knowledge. They have knowledge
21:16
unlimited knowledge. They have knowledge
21:16
unlimited knowledge. They have knowledge of anything. You can also tell them to
21:18
of anything. You can also tell them to
21:18
of anything. You can also tell them to go and learn these things. It can learn
21:21
go and learn these things. It can learn
21:21
go and learn these things. It can learn if it doesn't know all that. So that's
21:23
if it doesn't know all that. So that's
21:23
if it doesn't know all that. So that's how you have to think.
21:25
how you have to think.
21:25
how you have to think. Use the right keyword. Second point
21:27
Use the right keyword. Second point
21:27
Use the right keyword. Second point number two here is use the right
21:28
number two here is use the right
21:28
number two here is use the right keyword. How can I improve my code?
21:31
keyword. How can I improve my code?
21:31
keyword. How can I improve my code? That's one way to say. Other one is what
21:33
That's one way to say. Other one is what
21:33
That's one way to say. Other one is what are the effective strategies to optimize
21:35
are the effective strategies to optimize
21:35
are the effective strategies to optimize this Python function for performance and
21:38
this Python function for performance and
21:38
this Python function for performance and readability. So right. So this is I'm
21:40
readability. So right. So this is I'm
21:40
readability. So right. So this is I'm using the right keyword. What exactly
21:41
using the right keyword. What exactly
21:41
using the right keyword. What exactly I'm trying to do rather than just saying
21:44
I'm trying to do rather than just saying
21:44
I'm trying to do rather than just saying improvement. I'm saying look at this
21:46
improvement. I'm saying look at this
21:46
improvement. I'm saying look at this file first. You can also optimize or
21:48
file first. You can also optimize or
21:48
file first. You can also optimize or tell me how to optimize it and actually
21:50
tell me how to optimize it and actually
21:50
tell me how to optimize it and actually it will optimize for you.
21:54
it will optimize for you.
21:54
it will optimize for you. Third thing is third principle is
21:55
Third thing is third principle is
21:55
Third thing is third principle is provide context and context is very
21:58
provide context and context is very
21:58
provide context and context is very important. Let it know let AI know where
22:00
important. Let it know let AI know where
22:00
important. Let it know let AI know where you coming from what you're working on.
22:03
you coming from what you're working on.
22:03
you coming from what you're working on. So it knows a little bit background
22:06
So it knows a little bit background
22:06
So it knows a little bit background rather than just randomly asking the
22:08
rather than just randomly asking the
22:08
rather than just randomly asking the questions. Right? You can also say oh
22:11
questions. Right? You can also say oh
22:11
questions. Right? You can also say oh I'm working on this existing project
22:13
I'm working on this existing project
22:13
I'm working on this existing project that's written in C 8. Um I wrote that 3
22:17
that's written in C 8. Um I wrote that 3
22:17
that's written in C 8. Um I wrote that 3 years ago. It also uses ASP.NET.
22:20
years ago. It also uses ASP.NET.
22:20
years ago. It also uses ASP.NET. It doesn't uses ASP. Core. So when you
22:23
It doesn't uses ASP. Core. So when you
22:24
It doesn't uses ASP. Core. So when you are asking that specific question and
22:26
are asking that specific question and
22:26
are asking that specific question and providing the context and background now
22:28
providing the context and background now
22:28
providing the context and background now it start thinking like you and it start
22:31
it start thinking like you and it start
22:31
it start thinking like you and it start thinking like your project start
22:33
thinking like your project start
22:33
thinking like your project start thinking of your work right so another
22:35
thinking of your work right so another
22:35
thinking of your work right so another example is you can also give a
22:37
example is you can also give a
22:37
example is you can also give a background that act like a you know if
22:41
background that act like a you know if
22:41
background that act like a you know if you are let's say you are a project
22:43
you are let's say you are a project
22:43
you are let's say you are a project manager you can say act like a project
22:44
manager you can say act like a project
22:44
manager you can say act like a project manager who's a leading team of 12
22:46
manager who's a leading team of 12
22:46
manager who's a leading team of 12 people and blah blah blah so more you
22:49
people and blah blah blah so more you
22:49
people and blah blah blah so more you give the context and the background
22:51
give the context and the background
22:51
give the context and the background better AI would relate that to to its
22:56
better AI would relate that to to its
22:56
better AI would relate that to to its answers when it's crafting because it's
22:58
answers when it's crafting because it's
22:58
answers when it's crafting because it's going back to its you know large
22:59
going back to its you know large
23:00
going back to its you know large language model and going to look at all
23:01
language model and going to look at all
23:02
language model and going to look at all the examples it has related to those uh
23:05
the examples it has related to those uh
23:05
the examples it has related to those uh people project manager in this case.
23:08
people project manager in this case.
23:08
people project manager in this case. So these are six principles. I three
23:11
So these are six principles. I three
23:11
So these are six principles. I three principles is one is specific and
23:13
principles is one is specific and
23:13
principles is one is specific and direct. Second is use the right keyword.
23:15
direct. Second is use the right keyword.
23:15
direct. Second is use the right keyword. Third is provide context.
23:18
Third is provide context.
23:18
Third is provide context. Fourth is okay look my slides are messed
23:20
Fourth is okay look my slides are messed
23:20
Fourth is okay look my slides are messed up. Fourth is ask the format. You can
23:23
up. Fourth is ask the format. You can
23:23
up. Fourth is ask the format. You can also tell AI what format you want right?
23:26
also tell AI what format you want right?
23:26
also tell AI what format you want right? Uh you can just say oh explain this
23:28
Uh you can just say oh explain this
23:28
Uh you can just say oh explain this algorithm or you can say explain this or
23:31
algorithm or you can say explain this or
23:31
algorithm or you can say explain this or algorithm as a step step-by-step
23:33
algorithm as a step step-by-step
23:34
algorithm as a step step-by-step tutorial with code example and
23:35
tutorial with code example and
23:35
tutorial with code example and complexity analysis. So you can
23:37
complexity analysis. So you can
23:37
complexity analysis. So you can literally ask what kind of output you
23:39
literally ask what kind of output you
23:40
literally ask what kind of output you want from the AI
23:42
want from the AI
23:42
want from the AI and then the format. You can also tell
23:44
and then the format. You can also tell
23:44
and then the format. You can also tell as a tone right this break down complex
23:47
as a tone right this break down complex
23:47
as a tone right this break down complex question. So if you have this complex
23:48
question. So if you have this complex
23:48
question. So if you have this complex question try to break it down into a you
23:52
question try to break it down into a you
23:52
question try to break it down into a you know step by step so it can easily
23:54
know step by step so it can easily
23:54
know step by step so it can easily understand it's like sometimes you're
23:56
understand it's like sometimes you're
23:56
understand it's like sometimes you're talking to like you know assistant
23:59
talking to like you know assistant
23:59
talking to like you know assistant sometimes you can say this things in a
24:01
sometimes you can say this things in a
24:01
sometimes you can say this things in a big long sentence and or you can just
24:04
big long sentence and or you can just
24:04
big long sentence and or you can just say first do this after that I want you
24:06
say first do this after that I want you
24:06
say first do this after that I want you to do this then can you do that all
24:07
to do this then can you do that all
24:07
to do this then can you do that all those things so break down if you have a
24:10
those things so break down if you have a
24:10
those things so break down if you have a complex question break down into those
24:13
complex question break down into those
24:13
complex question break down into those so it can understand better so here's an
24:15
so it can understand better so here's an
24:15
so it can understand better so here's an example right build me a web application
24:17
example right build me a web application
24:17
example right build me a web application Web application can be very complex,
24:18
Web application can be very complex,
24:18
Web application can be very complex, right? What does that mean? It could be
24:21
right? What does that mean? It could be
24:21
right? What does that mean? It could be a web application doing different things
24:23
a web application doing different things
24:23
a web application doing different things or you can say first start a database
24:25
or you can say first start a database
24:25
or you can say first start a database schema with a user management table.
24:27
schema with a user management table.
24:28
schema with a user management table. What tables relationships would you
24:29
What tables relationships would you
24:29
What tables relationships would you recommend? You start with that, review
24:31
recommend? You start with that, review
24:31
recommend? You start with that, review that, change that, then you ask for
24:33
that, change that, then you ask for
24:33
that, change that, then you ask for another table. So you can do all that or
24:36
another table. So you can do all that or
24:36
another table. So you can do all that or you can just say build me a website with
24:38
you can just say build me a website with
24:38
you can just say build me a website with users in it. then it will just come up
24:40
users in it. then it will just come up
24:40
users in it. then it will just come up its own idea and come up with your own
24:42
its own idea and come up with your own
24:42
its own idea and come up with your own tables come up the fields and then it's
24:45
tables come up the fields and then it's
24:45
tables come up the fields and then it's just going to create and changing that
24:46
just going to create and changing that
24:46
just going to create and changing that is a hard part for you. So when you are
24:49
is a hard part for you. So when you are
24:49
is a hard part for you. So when you are building these complex products using
24:51
building these complex products using
24:51
building these complex products using chain AI. So by the way today as we are
24:54
chain AI. So by the way today as we are
24:54
chain AI. So by the way today as we are speaking right now you don't have to
24:56
speaking right now you don't have to
24:56
speaking right now you don't have to write a code to build a complex website.
24:58
write a code to build a complex website.
24:58
write a code to build a complex website. You can just use Genai tools and it will
25:01
You can just use Genai tools and it will
25:01
You can just use Genai tools and it will build the entire website. It will build
25:02
build the entire website. It will build
25:02
build the entire website. It will build your database. It will be your API. It
25:05
your database. It will be your API. It
25:05
your database. It will be your API. It will build your front end. It will build
25:06
will build your front end. It will build
25:06
will build your front end. It will build your business logic without writing
25:09
your business logic without writing
25:09
your business logic without writing code. It can all do that but you have to
25:11
code. It can all do that but you have to
25:11
code. It can all do that but you have to just know how to talk to it and how to
25:13
just know how to talk to it and how to
25:13
just know how to talk to it and how to do this step by step. you know like how
25:16
do this step by step. you know like how
25:16
do this step by step. you know like how you would do with any assistant
25:17
you would do with any assistant
25:18
you would do with any assistant programmers and and DBAs and um you know
25:21
programmers and and DBAs and um you know
25:21
programmers and and DBAs and um you know back-end developers. This is kind of you
25:23
back-end developers. This is kind of you
25:23
back-end developers. This is kind of you can tell AI sometimes become a DBA build
25:26
can tell AI sometimes become a DBA build
25:26
can tell AI sometimes become a DBA build your database it can also become your
25:28
your database it can also become your
25:28
your database it can also become your backend developer write your APIs and
25:31
backend developer write your APIs and
25:31
backend developer write your APIs and backend logic it can also do your build
25:34
backend logic it can also do your build
25:34
backend logic it can also do your build your front end. So now you have all
25:36
your front end. So now you have all
25:36
your front end. So now you have all these developers at your disposal they
25:39
these developers at your disposal they
25:39
these developers at your disposal they can AI can act like whatever you want it
25:41
can AI can act like whatever you want it
25:41
can AI can act like whatever you want it to be.
25:43
to be.
25:43
to be. Um and then last thing is set the right
25:44
Um and then last thing is set the right
25:44
Um and then last thing is set the right persona and tone and those are
25:46
persona and tone and those are
25:46
persona and tone and those are important. Persona means when you are
25:48
important. Persona means when you are
25:48
important. Persona means when you are asking a prompt you can specifically
25:51
asking a prompt you can specifically
25:51
asking a prompt you can specifically unless it's a general prompt you will
25:53
unless it's a general prompt you will
25:53
unless it's a general prompt you will say oh I'm a business manager running a
25:57
say oh I'm a business manager running a
25:57
say oh I'm a business manager running a car agency we sell new cars. So now AI
26:02
car agency we sell new cars. So now AI
26:02
car agency we sell new cars. So now AI is going to write the C output output
26:05
is going to write the C output output
26:05
is going to write the C output output will come that any in anybody who's car
26:09
will come that any in anybody who's car
26:09
will come that any in anybody who's car you know manager and sitting and blah
26:12
you know manager and sitting and blah
26:12
you know manager and sitting and blah blah that's how in case say you are a
26:14
blah that's how in case say you are a
26:14
blah that's how in case say you are a software architect then you will say act
26:16
software architect then you will say act
26:16
software architect then you will say act as a so as a senior software architect
26:18
as a so as a senior software architect
26:18
as a so as a senior software architect if you are a student you can say act as
26:20
if you are a student you can say act as
26:20
if you are a student you can say act as a student right and create a letter um
26:24
a student right and create a letter um
26:24
a student right and create a letter um internship letter you can do this is how
26:26
internship letter you can do this is how
26:26
internship letter you can do this is how you to explain and you can also say tone
26:29
you to explain and you can also say tone
26:29
you to explain and you can also say tone if tone doesn't apply So you have to
26:31
if tone doesn't apply So you have to
26:31
if tone doesn't apply So you have to think guy the six all these things I'm
26:33
think guy the six all these things I'm
26:33
think guy the six all these things I'm saying some may not apply to other some
26:36
saying some may not apply to other some
26:36
saying some may not apply to other some questions so you don't have to force it
26:38
questions so you don't have to force it
26:38
questions so you don't have to force it okay so if you're asking what is capital
26:41
okay so if you're asking what is capital
26:41
okay so if you're asking what is capital of India we just say it's a capital of
26:42
of India we just say it's a capital of
26:42
of India we just say it's a capital of India simple question you don't need
26:44
India simple question you don't need
26:44
India simple question you don't need tone you don't need all that right so
26:46
tone you don't need all that right so
26:46
tone you don't need all that right so you have to always make that decision
26:48
you have to always make that decision
26:48
you have to always make that decision not all these applies all these
26:50
not all these applies all these
26:50
not all these applies all these principles apply in every prompt
26:54
principles apply in every prompt
26:54
principles apply in every prompt okay hopefully you know you guys learned
26:55
okay hopefully you know you guys learned
26:55
okay hopefully you know you guys learned something so quickly what are the
26:58
something so quickly what are the
26:58
something so quickly what are the different core principles of good
26:59
different core principles of good
26:59
different core principles of good prompting one is be specific and direct.
27:02
prompting one is be specific and direct.
27:02
prompting one is be specific and direct. Use the right keywords. Provide context.
27:05
Use the right keywords. Provide context.
27:05
Use the right keywords. Provide context. Um, explain the format you want. Break
27:08
Um, explain the format you want. Break
27:08
Um, explain the format you want. Break down complex questions to smaller tasks.
27:11
down complex questions to smaller tasks.
27:11
down complex questions to smaller tasks. And use set the right personal tone. So,
27:16
And use set the right personal tone. So,
27:16
And use set the right personal tone. So, I can show you live demo now, but I
27:19
I can show you live demo now, but I
27:19
I can show you live demo now, but I think most of you already know how to do
27:21
think most of you already know how to do
27:21
think most of you already know how to do this. You can literally go and type in
27:23
this. You can literally go and type in
27:23
this. You can literally go and type in chat GPT. Um, because I'm going to
27:26
chat GPT. Um, because I'm going to
27:26
chat GPT. Um, because I'm going to otherwise use the screen and you say
27:27
otherwise use the screen and you say
27:27
otherwise use the screen and you say write an email. It will write just
27:29
write an email. It will write just
27:29
write an email. It will write just whatever email then or you can say
27:31
whatever email then or you can say
27:31
whatever email then or you can say specific write an email do this do this
27:35
specific write an email do this do this
27:35
specific write an email do this do this if you want maybe I can quickly show you
27:37
if you want maybe I can quickly show you
27:37
if you want maybe I can quickly show you wonder if I can share my screen
27:41
wonder if I can share my screen
27:41
wonder if I can share my screen uh
27:46
share my screen let's see share screen
27:50
share my screen let's see share screen
27:50
share my screen let's see share screen I'll say entire window
27:54
I'll say entire window
27:54
I'll say entire window it's okay I should be sharing my screen
27:56
it's okay I should be sharing my screen
27:56
it's okay I should be sharing my screen now so let me Click open here
28:01
now so let me Click open here
28:01
now so let me Click open here it will chat GPT
28:04
it will chat GPT
28:04
it will chat GPT and I'll say write an email right we'll
28:06
and I'll say write an email right we'll
28:06
and I'll say write an email right we'll just take an example it will just write
28:08
just take an example it will just write
28:08
just take an example it will just write whatever email right okay now it will
28:11
whatever email right okay now it will
28:11
whatever email right okay now it will first ask all these things good is it's
28:13
first ask all these things good is it's
28:13
first ask all these things good is it's getting smarter these days it's already
28:14
getting smarter these days it's already
28:14
getting smarter these days it's already asking so now I can say
28:18
asking so now I can say
28:18
asking so now I can say I am a bank manager
28:22
I am a bank manager
28:22
I am a bank manager writing an email to a customer who is
28:26
writing an email to a customer who is
28:26
writing an email to a customer who is late on his payments in monthly payment.
28:34
Be friendly
28:37
Be friendly
28:37
Be friendly and also let him know if there are
28:42
and also let him know if there are
28:42
and also let him know if there are any problems.
28:46
any problems.
28:46
any problems. We can help with the payment
28:51
and even remove this month
28:56
and even remove this month
28:56
and even remove this month from his notice. Whatever you get the
28:59
from his notice. Whatever you get the
28:59
from his notice. Whatever you get the idea. Now you can also say what is the
29:02
idea. Now you can also say what is the
29:02
idea. Now you can also say what is the bank? What is he like this exactly what
29:04
bank? What is he like this exactly what
29:04
bank? What is he like this exactly what says right? So it wrote a very detailed
29:06
says right? So it wrote a very detailed
29:06
says right? So it wrote a very detailed output and more prompt engineering
29:10
output and more prompt engineering
29:10
output and more prompt engineering something you have to practice more more
29:12
something you have to practice more more
29:12
something you have to practice more more you practice next time better you get at
29:15
you practice next time better you get at
29:15
you practice next time better you get at it. Right now you may say okay I should
29:17
it. Right now you may say okay I should
29:17
it. Right now you may say okay I should have done that. So that's called you
29:19
have done that. So that's called you
29:19
have done that. So that's called you know iteration. Iteration is also a big
29:22
know iteration. Iteration is also a big
29:22
know iteration. Iteration is also a big part of prompt engineering writing good
29:24
part of prompt engineering writing good
29:24
part of prompt engineering writing good prompts you can learn from that. But
29:26
prompts you can learn from that. But
29:26
prompts you can learn from that. But it's better to come up with this in the
29:28
it's better to come up with this in the
29:28
it's better to come up with this in the first because now you are only sending
29:31
first because now you are only sending
29:31
first because now you are only sending one prompt to AI. If you send multiple
29:34
one prompt to AI. If you send multiple
29:34
one prompt to AI. If you send multiple prompts and keep sending them back and
29:36
prompts and keep sending them back and
29:36
prompts and keep sending them back and forth, not only is going to take you
29:38
forth, not only is going to take you
29:38
forth, not only is going to take you more time, but it's going to cost a lot
29:41
more time, but it's going to cost a lot
29:41
more time, but it's going to cost a lot behind the scene. Every time you are
29:43
behind the scene. Every time you are
29:43
behind the scene. Every time you are sending this input text, there's
29:45
sending this input text, there's
29:45
sending this input text, there's machines running, a lot of computers
29:47
machines running, a lot of computers
29:47
machines running, a lot of computers running behind the scene, right? So, you
29:49
running behind the scene, right? So, you
29:49
running behind the scene, right? So, you don't want to waste the processing power
29:51
don't want to waste the processing power
29:51
don't want to waste the processing power and time and also gets costly. So, this
29:55
and time and also gets costly. So, this
29:55
and time and also gets costly. So, this gives you an idea.
29:57
gives you an idea.
29:57
gives you an idea. Um, so let's get back to our screen now.
30:01
Um, so let's get back to our screen now.
30:01
Um, so let's get back to our screen now. Um
30:04
before uh we go to the questions, looks
30:06
before uh we go to the questions, looks
30:06
before uh we go to the questions, looks like there's a lot of questions coming
30:08
like there's a lot of questions coming
30:08
like there's a lot of questions coming up. Um we will quickly talk about again
30:11
up. Um we will quickly talk about again
30:11
up. Um we will quickly talk about again you um thank you again for shar token.
30:14
you um thank you again for shar token.
30:14
you um thank you again for shar token. Check it out. Don't forget guys uh this
30:17
Check it out. Don't forget guys uh this
30:17
Check it out. Don't forget guys uh this is something we are building as a reward
30:19
is something we are building as a reward
30:19
is something we are building as a reward for C# corner. So man that that means if
30:22
for C# corner. So man that that means if
30:22
for C# corner. So man that that means if you are watching this show in coming
30:24
you are watching this show in coming
30:24
you are watching this show in coming time you will actually get shop token
30:26
time you will actually get shop token
30:26
time you will actually get shop token rewards. If you are contributing
30:29
rewards. If you are contributing
30:29
rewards. If you are contributing anything on C# corner, you will get
30:30
anything on C# corner, you will get
30:30
anything on C# corner, you will get these rewards. So we are working on
30:32
these rewards. So we are working on
30:32
these rewards. So we are working on that. Check it out.
30:34
that. Check it out.
30:34
that. Check it out. Then I have two more announcements for
30:37
Then I have two more announcements for
30:37
Then I have two more announcements for coming weeks. So next week I'm going to
30:40
coming weeks. So next week I'm going to
30:40
coming weeks. So next week I'm going to focus on mastering AI prompts. I have
30:42
focus on mastering AI prompts. I have
30:42
focus on mastering AI prompts. I have created this framework called recast
30:45
created this framework called recast
30:45
created this framework called recast framework. It's simple way for you to
30:47
framework. It's simple way for you to
30:47
framework. It's simple way for you to learn and remember the prompting. Okay,
30:51
learn and remember the prompting. Okay,
30:51
learn and remember the prompting. Okay, it's called recast framework. It you
30:52
it's called recast framework. It you
30:52
it's called recast framework. It you know re R is for something, E for
30:54
know re R is for something, E for
30:54
know re R is for something, E for something all those content. So that's
30:56
something all those content. So that's
30:56
something all those content. So that's where I will talk. So next week we will
30:59
where I will talk. So next week we will
30:59
where I will talk. So next week we will dig down into more examples hands-on
31:02
dig down into more examples hands-on
31:02
dig down into more examples hands-on different kind of complex problems.
31:04
different kind of complex problems.
31:04
different kind of complex problems. We'll do all that work and look at the
31:06
We'll do all that work and look at the
31:06
We'll do all that work and look at the different outputs how how different they
31:08
different outputs how how different they
31:08
different outputs how how different they are. That will be next week and then
31:11
are. That will be next week and then
31:11
are. That will be next week and then following week. So this is my recast
31:13
following week. So this is my recast
31:13
following week. So this is my recast framework quickly just give you like the
31:14
framework quickly just give you like the
31:14
framework quickly just give you like the role you have to create as a role who
31:17
role you have to create as a role who
31:17
role you have to create as a role who you role example context actions
31:19
you role example context actions
31:19
you role example context actions structure tone. So I call it recast
31:22
structure tone. So I call it recast
31:22
structure tone. So I call it recast framework. It's the same thing we should
31:24
framework. It's the same thing we should
31:24
framework. It's the same thing we should talk but it's kind of puts you in
31:25
talk but it's kind of puts you in
31:25
talk but it's kind of puts you in framework and structure and you'll
31:27
framework and structure and you'll
31:27
framework and structure and you'll remember this easily every time you are
31:29
remember this easily every time you are
31:29
remember this easily every time you are building these you know complex
31:31
building these you know complex
31:31
building these you know complex applications and then following week we
31:34
applications and then following week we
31:34
applications and then following week we will talk about advanced prompt
31:35
will talk about advanced prompt
31:36
will talk about advanced prompt engineering techniques. So there's one
31:37
engineering techniques. So there's one
31:37
engineering techniques. So there's one shot few short there's multiple advanced
31:40
shot few short there's multiple advanced
31:40
shot few short there's multiple advanced techniques as well that will come and
31:42
techniques as well that will come and
31:42
techniques as well that will come and then following week I'll also then focus
31:44
then following week I'll also then focus
31:44
then following week I'll also then focus on LLM inside we'll go inside the LLMs
31:48
on LLM inside we'll go inside the LLMs
31:48
on LLM inside we'll go inside the LLMs how LM work how they store the data
31:51
how LM work how they store the data
31:51
how LM work how they store the data relationships and all and how you can
31:53
relationships and all and how you can
31:53
relationships and all and how you can even start building your LLM so we'll go
31:55
even start building your LLM so we'll go
31:55
even start building your LLM so we'll go more into that coming weeks with that
31:58
more into that coming weeks with that
31:58
more into that coming weeks with that said guys let's take some questions now
32:00
said guys let's take some questions now
32:00
said guys let's take some questions now I think we are good on timing went a
32:03
I think we are good on timing went a
32:03
I think we are good on timing went a little bit longer than I thought but
32:05
little bit longer than I thought but
32:05
little bit longer than I thought but yeah let's take some questions questions
32:06
yeah let's take some questions questions
32:06
yeah let's take some questions questions and then um
32:10
and then um
32:10
and then um looks like a lot of questions we got. Um
32:14
looks like a lot of questions we got. Um
32:14
looks like a lot of questions we got. Um let me make the screen big for a second.
32:17
let me make the screen big for a second.
32:17
let me make the screen big for a second. Um how can students or beginner build a
32:20
Um how can students or beginner build a
32:20
Um how can students or beginner build a carry on prompt engineering? Are there
32:21
carry on prompt engineering? Are there
32:21
carry on prompt engineering? Are there any specific tools or platform we should
32:23
any specific tools or platform we should
32:24
any specific tools or platform we should master? Great question Vidwan. I think
32:27
master? Great question Vidwan. I think
32:27
master? Great question Vidwan. I think the easiest way to learn prompt
32:30
the easiest way to learn prompt
32:30
the easiest way to learn prompt engineering is start with a simple
32:31
engineering is start with a simple
32:31
engineering is start with a simple training uh or there's courses out there
32:34
training uh or there's courses out there
32:34
training uh or there's courses out there online as well. and watch this show. You
32:36
online as well. and watch this show. You
32:36
online as well. and watch this show. You already watched this show. You hopefully
32:38
already watched this show. You hopefully
32:38
already watched this show. You hopefully learn little bit. But the future is
32:40
learn little bit. But the future is
32:40
learn little bit. But the future is prompt engineering. Regardless of what
32:42
prompt engineering. Regardless of what
32:42
prompt engineering. Regardless of what you are, you can be a computer science
32:45
you are, you can be a computer science
32:45
you are, you can be a computer science major, you can be a software engineer,
32:48
major, you can be a software engineer,
32:48
major, you can be a software engineer, you can be a business owner, you can be
32:50
you can be a business owner, you can be
32:50
you can be a business owner, you can be anybody. Prompting is everybody should
32:52
anybody. Prompting is everybody should
32:52
anybody. Prompting is everybody should know. And if you are technical, you
32:54
know. And if you are technical, you
32:54
know. And if you are technical, you should be more in in in depth into
32:56
should be more in in in depth into
32:56
should be more in in in depth into prompt engineering. So we also on C#
32:59
prompt engineering. So we also on C#
32:59
prompt engineering. So we also on C# corner, we are launching a course for
33:01
corner, we are launching a course for
33:01
corner, we are launching a course for you guys. Especially there are two
33:03
you guys. Especially there are two
33:03
you guys. Especially there are two courses. One is for students, another
33:05
courses. One is for students, another
33:05
courses. One is for students, another one is for experienced developers and
33:07
one is for experienced developers and
33:07
one is for experienced developers and one is for engineering teams and these
33:10
one is for engineering teams and these
33:10
one is for engineering teams and these courses will be announced next week on
33:12
courses will be announced next week on
33:12
courses will be announced next week on C# corner. So you guys have already
33:15
C# corner. So you guys have already
33:15
C# corner. So you guys have already registered. We will send you the details
33:16
registered. We will send you the details
33:16
registered. We will send you the details if you want to register on those
33:18
if you want to register on those
33:18
if you want to register on those courses.
33:19
courses.
33:19
courses. Um let's take another question.
33:24
Um let's take another question.
33:24
Um let's take another question. Um how these companies using prompting
33:27
Um how these companies using prompting
33:27
Um how these companies using prompting in their workflow today? So this is
33:29
in their workflow today? So this is
33:29
in their workflow today? So this is great question Kushi. Um, every company
33:32
great question Kushi. Um, every company
33:32
great question Kushi. Um, every company is now integrating AI into their
33:36
is now integrating AI into their
33:36
is now integrating AI into their business processes. And when you
33:38
business processes. And when you
33:38
business processes. And when you integrate AI, all they they their team
33:42
integrate AI, all they they their team
33:42
integrate AI, all they they their team needs to know how to build these prompts
33:44
needs to know how to build these prompts
33:44
needs to know how to build these prompts and how to craft the prompt. Give you an
33:47
and how to craft the prompt. Give you an
33:47
and how to craft the prompt. Give you an example. I have a friend, you know, at
33:50
example. I have a friend, you know, at
33:50
example. I have a friend, you know, at Amazon. They said all they talk about
33:53
Amazon. They said all they talk about
33:53
Amazon. They said all they talk about all day Jai and Jai relies on prompt
33:56
all day Jai and Jai relies on prompt
33:56
all day Jai and Jai relies on prompt engineering. Prompt engineering the key
33:59
engineering. Prompt engineering the key
33:59
engineering. Prompt engineering the key to work and get more productive. It's
34:01
to work and get more productive. It's
34:02
to work and get more productive. It's like programming right? If you are
34:03
like programming right? If you are
34:03
like programming right? If you are getting hired by a company, if you are a
34:06
getting hired by a company, if you are a
34:06
getting hired by a company, if you are a good programmer and you know a lot of
34:07
good programmer and you know a lot of
34:07
good programmer and you know a lot of commands, you can build things faster
34:09
commands, you can build things faster
34:09
commands, you can build things faster rather than learning and copying and
34:11
rather than learning and copying and
34:11
rather than learning and copying and searching and all. Same thing with
34:13
searching and all. Same thing with
34:13
searching and all. Same thing with prompt engineering. More gen more you
34:15
prompt engineering. More gen more you
34:15
prompt engineering. More gen more you want to get out of jai and faster better
34:18
want to get out of jai and faster better
34:18
want to get out of jai and faster better you are at prompt engineering more you
34:20
you are at prompt engineering more you
34:20
you are at prompt engineering more you will get there more productive you'll
34:21
will get there more productive you'll
34:21
will get there more productive you'll become.
34:24
become.
34:24
become. Great questions guys keep coming those
34:25
Great questions guys keep coming those
34:26
Great questions guys keep coming those questions. Um, Sartuk has a question. As
34:29
questions. Um, Sartuk has a question. As
34:29
questions. Um, Sartuk has a question. As large language model keeps evolving, do
34:31
large language model keeps evolving, do
34:31
large language model keeps evolving, do you think prompt engineering will become
34:33
you think prompt engineering will become
34:33
you think prompt engineering will become essential for everybody working with AI
34:35
essential for everybody working with AI
34:35
essential for everybody working with AI tools and how can beginner short start
34:38
tools and how can beginner short start
34:38
tools and how can beginner short start learning and write better prompts in
34:39
learning and write better prompts in
34:39
learning and write better prompts in LLM? Yeah, that's yeah 100%. 100% prompt
34:43
LLM? Yeah, that's yeah 100%. 100% prompt
34:43
LLM? Yeah, that's yeah 100%. 100% prompt engineering will become a skill
34:45
engineering will become a skill
34:45
engineering will become a skill everybody would need um especially in
34:47
everybody would need um especially in
34:47
everybody would need um especially in the tech tech space. Okay. Um the way is
34:51
the tech tech space. Okay. Um the way is
34:51
the tech tech space. Okay. Um the way is better prompts is again as I said launch
34:54
better prompts is again as I said launch
34:54
better prompts is again as I said launch watch these shows like this like my
34:56
watch these shows like this like my
34:56
watch these shows like this like my framework will explain more and then we
34:58
framework will explain more and then we
34:58
framework will explain more and then we are also launching a training as well
35:00
are also launching a training as well
35:00
are also launching a training as well and there's YouTube has lot of stuff
35:02
and there's YouTube has lot of stuff
35:02
and there's YouTube has lot of stuff sometimes hard to find things on YouTube
35:05
sometimes hard to find things on YouTube
35:05
sometimes hard to find things on YouTube so that's why we are creating these
35:06
so that's why we are creating these
35:06
so that's why we are creating these specific
35:08
specific
35:08
specific courses where anybody can go and learn
35:11
courses where anybody can go and learn
35:11
courses where anybody can go and learn this
35:13
this
35:13
this all right let's take more questions uh
35:15
all right let's take more questions uh
35:15
all right let's take more questions uh the question if Chad eventually gives
35:18
the question if Chad eventually gives
35:18
the question if Chad eventually gives the right answer after a few try then
35:20
the right answer after a few try then
35:20
the right answer after a few try then why do we need prompt engineering isn't
35:23
why do we need prompt engineering isn't
35:23
why do we need prompt engineering isn't that just trial and error so that is
35:26
that just trial and error so that is
35:26
that just trial and error so that is correct it will give you right answers
35:29
correct it will give you right answers
35:29
correct it will give you right answers but again as I said more prompt you give
35:32
but again as I said more prompt you give
35:32
but again as I said more prompt you give it to more costly become okay so after
35:36
it to more costly become okay so after
35:36
it to more costly become okay so after sometimes what will happen is if you ask
35:39
sometimes what will happen is if you ask
35:39
sometimes what will happen is if you ask this question 100 times chat it will say
35:41
this question 100 times chat it will say
35:41
this question 100 times chat it will say that you have reached your limits now
35:43
that you have reached your limits now
35:43
that you have reached your limits now you have to pay me because right now
35:46
you have to pay me because right now
35:46
you have to pay me because right now everything is free but there companies
35:48
everything is free but there companies
35:48
everything is free but there companies are not going to continue to give people
35:50
are not going to continue to give people
35:50
are not going to continue to give people free stuff. You know when a lot of
35:53
free stuff. You know when a lot of
35:53
free stuff. You know when a lot of things launch in the beginning they are
35:54
things launch in the beginning they are
35:54
things launch in the beginning they are free because end of the day it's costing
35:57
free because end of the day it's costing
35:57
free because end of the day it's costing company money. Every time you ask a
35:58
company money. Every time you ask a
35:58
company money. Every time you ask a question it's costing them money. So
36:01
question it's costing them money. So
36:01
question it's costing them money. So less questions you ask and how do you
36:04
less questions you ask and how do you
36:04
less questions you ask and how do you become less question is more it's it's
36:06
become less question is more it's it's
36:06
become less question is more it's it's like different less stake less mistake
36:08
like different less stake less mistake
36:08
like different less stake less mistake it will make. It's like cook like you
36:11
it will make. It's like cook like you
36:11
it will make. It's like cook like you have a cook and you tell them just say
36:14
have a cook and you tell them just say
36:14
have a cook and you tell them just say oh can you go and cook this thing
36:16
oh can you go and cook this thing
36:16
oh can you go and cook this thing they'll cook like this is not good redo
36:18
they'll cook like this is not good redo
36:18
they'll cook like this is not good redo it again. So it's like process it takes
36:20
it again. So it's like process it takes
36:20
it again. So it's like process it takes time and they are spending resources you
36:22
time and they are spending resources you
36:22
time and they are spending resources you may not realize right now if when you
36:24
may not realize right now if when you
36:24
may not realize right now if when you use chat GPT you may not realize but
36:26
use chat GPT you may not realize but
36:26
use chat GPT you may not realize but behind the scene there's a lot of
36:28
behind the scene there's a lot of
36:28
behind the scene there's a lot of process goes on when you ask a question
36:31
process goes on when you ask a question
36:31
process goes on when you ask a question more questions you ask more cost it's
36:33
more questions you ask more cost it's
36:34
more questions you ask more cost it's going to be and eventually the like
36:36
going to be and eventually the like
36:36
going to be and eventually the like right now I use copilot to make images
36:39
right now I use copilot to make images
36:39
right now I use copilot to make images okay I go to copilot you can make images
36:42
okay I go to copilot you can make images
36:42
okay I go to copilot you can make images but it will let you only create few free
36:44
but it will let you only create few free
36:44
but it will let you only create few free images per day after three four five
36:47
images per day after three four five
36:47
images per day after three four five images it says Oh, your limit per day is
36:49
images it says Oh, your limit per day is
36:49
images it says Oh, your limit per day is gone. Now you have to become a paid
36:51
gone. Now you have to become a paid
36:51
gone. Now you have to become a paid member. So now I have to pay for that.
36:54
member. So now I have to pay for that.
36:54
member. So now I have to pay for that. So that's the that's the reason that's
36:56
So that's the that's the reason that's
36:56
So that's the that's the reason that's the reason and companies don't have p
36:58
the reason and companies don't have p
36:58
the reason and companies don't have p time to mess around with different pawn
37:00
time to mess around with different pawn
37:00
time to mess around with different pawn because now you are taking more time and
37:02
because now you are taking more time and
37:02
because now you are taking more time and company will say why is taking so much
37:04
company will say why is taking so much
37:04
company will say why is taking so much time it should be done in like few
37:05
time it should be done in like few
37:06
time it should be done in like few minutes. So funny thing is you know six
37:09
minutes. So funny thing is you know six
37:09
minutes. So funny thing is you know six months or a year ago when there was a
37:11
months or a year ago when there was a
37:11
months or a year ago when there was a new project you know coming to companies
37:14
new project you know coming to companies
37:14
new project you know coming to companies they will say oh can you estimate me and
37:16
they will say oh can you estimate me and
37:16
they will say oh can you estimate me and estimates will be it will take us that
37:18
estimates will be it will take us that
37:18
estimates will be it will take us that many months that many weeks now new
37:21
many months that many weeks now new
37:21
many months that many weeks now new requirements come it says how many hours
37:23
requirements come it says how many hours
37:23
requirements come it says how many hours and minutes it will take you using jai
37:26
and minutes it will take you using jai
37:26
and minutes it will take you using jai so that's what matters time is the
37:28
so that's what matters time is the
37:28
so that's what matters time is the important part here and because time and
37:31
important part here and because time and
37:31
important part here and because time and processing that's why ch prompt
37:34
processing that's why ch prompt
37:34
processing that's why ch prompt engineering becomes very important skill
37:38
Okay, next question. Um,
37:43
Okay, next question. Um,
37:43
Okay, next question. Um, if I close the tab and clear chat, the
37:47
if I close the tab and clear chat, the
37:47
if I close the tab and clear chat, the chat GP keep what I said. It will
37:49
chat GP keep what I said. It will
37:49
chat GP keep what I said. It will depends on that. I think it depends on
37:51
depends on that. I think it depends on
37:51
depends on that. I think it depends on if you tell it to remember, it can
37:53
if you tell it to remember, it can
37:54
if you tell it to remember, it can remember. There's also settings, can
37:55
remember. There's also settings, can
37:55
remember. There's also settings, can keep your, if you go to chat GPT, it can
37:59
keep your, if you go to chat GPT, it can
37:59
keep your, if you go to chat GPT, it can keep your u context, it can keep your
38:01
keep your u context, it can keep your
38:01
keep your u context, it can keep your history as well. So, yeah.
38:09
So question is next you ask what daily
38:11
So question is next you ask what daily
38:11
So question is next you ask what daily habits or exercise would you recommend
38:13
habits or exercise would you recommend
38:13
habits or exercise would you recommend to improve our prompt in writing skills.
38:17
to improve our prompt in writing skills.
38:17
to improve our prompt in writing skills. It's it's simple it's practice learning
38:20
It's it's simple it's practice learning
38:20
It's it's simple it's practice learning more watching shows like this watching
38:22
more watching shows like this watching
38:22
more watching shows like this watching more things on YouTube or anywhere you
38:25
more things on YouTube or anywhere you
38:25
more things on YouTube or anywhere you can find and it's really putting time in
38:28
can find and it's really putting time in
38:28
can find and it's really putting time in there. There's no secret formula, right?
38:30
there. There's no secret formula, right?
38:30
there. There's no secret formula, right? Only secret formula is keep doing this
38:32
Only secret formula is keep doing this
38:32
Only secret formula is keep doing this more, learning more, learning more
38:34
more, learning more, learning more
38:34
more, learning more, learning more words. Another thing you can do is you
38:37
words. Another thing you can do is you
38:37
words. Another thing you can do is you can look at some of these uh popular
38:40
can look at some of these uh popular
38:40
can look at some of these uh popular chat prompts how people are crafting
38:43
chat prompts how people are crafting
38:43
chat prompts how people are crafting these prompts. So some of some of these
38:46
these prompts. So some of some of these
38:46
these prompts. So some of some of these talking to more talking to to more and
38:48
talking to more talking to to more and
38:48
talking to more talking to to more and see if they are giving you the right
38:49
see if they are giving you the right
38:49
see if they are giving you the right output you want and see if you go to
38:53
output you want and see if you go to
38:53
output you want and see if you go to bunch of these platforms now will show
38:55
bunch of these platforms now will show
38:55
bunch of these platforms now will show you existing prompt and what people are
38:57
you existing prompt and what people are
38:57
you existing prompt and what people are creating. So chat GP now have a
39:00
creating. So chat GP now have a
39:00
creating. So chat GP now have a something called Sora right this openai
39:02
something called Sora right this openai
39:02
something called Sora right this openai has a platform Sora Sora creates images
39:05
has a platform Sora Sora creates images
39:05
has a platform Sora Sora creates images and videos for you. So you go to Sora
39:08
and videos for you. So you go to Sora
39:08
and videos for you. So you go to Sora you can see all the sample prompts and
39:11
you can see all the sample prompts and
39:11
you can see all the sample prompts and see what output they have created.
39:14
see what output they have created.
39:14
see what output they have created. Sometimes do you see you know you are on
39:15
Sometimes do you see you know you are on
39:15
Sometimes do you see you know you are on Instagram sometimes you're on Instagram
39:17
Instagram sometimes you're on Instagram
39:17
Instagram sometimes you're on Instagram and you'll see oh somebody has created
39:19
and you'll see oh somebody has created
39:19
and you'll see oh somebody has created amazing videos using AI somebody's bird
39:23
amazing videos using AI somebody's bird
39:23
amazing videos using AI somebody's bird is flying all those you can even input
39:26
is flying all those you can even input
39:26
is flying all those you can even input that in some of these tools and say can
39:30
that in some of these tools and say can
39:30
that in some of these tools and say can you create a prompt for me and then you
39:32
you create a prompt for me and then you
39:32
you create a prompt for me and then you learn from that what kind of prompt
39:34
learn from that what kind of prompt
39:34
learn from that what kind of prompt they're creating. So that's how you can
39:36
they're creating. So that's how you can
39:36
they're creating. So that's how you can do uh you can get better at it. It's
39:39
do uh you can get better at it. It's
39:39
do uh you can get better at it. It's really spending time more time on it.
39:44
And next question is uh let's take
39:47
And next question is uh let's take
39:47
And next question is uh let's take another question here guys. Uh how do I
39:50
another question here guys. Uh how do I
39:50
another question here guys. Uh how do I make the best use of features like deep
39:53
make the best use of features like deep
39:53
make the best use of features like deep search and chat GPT for task like
39:55
search and chat GPT for task like
39:55
search and chat GPT for task like project research which can only be used
39:58
project research which can only be used
39:58
project research which can only be used 5x a week I believe. Yes. So there's
40:01
5x a week I believe. Yes. So there's
40:01
5x a week I believe. Yes. So there's limitations right the deep research has
40:03
limitations right the deep research has
40:03
limitations right the deep research has limitations
40:04
limitations
40:04
limitations and in the sense and the reason it has
40:06
and in the sense and the reason it has
40:06
and in the sense and the reason it has limitations keep in mind is that every
40:08
limitations keep in mind is that every
40:08
limitations keep in mind is that every time you ask a question deep research
40:10
time you ask a question deep research
40:10
time you ask a question deep research mode it is a lot of more work research
40:13
mode it is a lot of more work research
40:13
mode it is a lot of more work research is doing for you the answer it doesn't
40:15
is doing for you the answer it doesn't
40:15
is doing for you the answer it doesn't have it is going to do go to millions of
40:18
have it is going to do go to millions of
40:18
have it is going to do go to millions of you know millions of parameters go into
40:20
you know millions of parameters go into
40:20
you know millions of parameters go into the neural network look for more things
40:22
the neural network look for more things
40:22
the neural network look for more things search from data it has to do more work
40:25
search from data it has to do more work
40:25
search from data it has to do more work so rather than and right now they're
40:27
so rather than and right now they're
40:27
so rather than and right now they're giving you all this free but in coming
40:29
giving you all this free but in coming
40:29
giving you all this free but in coming time Everything will become paid. I can
40:32
time Everything will become paid. I can
40:32
time Everything will become paid. I can tell you for that. Remember when Gmail
40:34
tell you for that. Remember when Gmail
40:34
tell you for that. Remember when Gmail used to be free. You can just get a
40:36
used to be free. You can just get a
40:36
used to be free. You can just get a Gmail. You can do so much free space and
40:39
Gmail. You can do so much free space and
40:39
Gmail. You can do so much free space and then over the time they're like okay now
40:40
then over the time they're like okay now
40:40
then over the time they're like okay now you have to pay. Now they go you got
40:42
you have to pay. Now they go you got
40:42
you have to pay. Now they go you got stuck there. So this will happen also
40:45
stuck there. So this will happen also
40:45
stuck there. So this will happen also with these tools. So answer to your
40:48
with these tools. So answer to your
40:48
with these tools. So answer to your upshake to your question is you have to
40:51
upshake to your question is you have to
40:51
upshake to your question is you have to think like do I really need a deep
40:52
think like do I really need a deep
40:52
think like do I really need a deep research for that deep search for that
40:55
research for that deep search for that
40:55
research for that deep search for that because something can be just general if
40:57
because something can be just general if
40:57
because something can be just general if it's a general question and general
40:58
it's a general question and general
40:58
it's a general question and general answer you're fine with that but
41:00
answer you're fine with that but
41:00
answer you're fine with that but something that cannot be find more
41:02
something that cannot be find more
41:02
something that cannot be find more specifically like you're doing a
41:04
specifically like you're doing a
41:04
specifically like you're doing a research that's not available everywhere
41:06
research that's not available everywhere
41:06
research that's not available everywhere else then you want to use it so you got
41:08
else then you want to use it so you got
41:08
else then you want to use it so you got to use smart with that
41:12
um yeah so yeah chit pro is it's not a I
41:16
um yeah so yeah chit pro is it's not a I
41:16
um yeah so yeah chit pro is it's not a I I wouldn't say yes answer is yes. It's I
41:18
I wouldn't say yes answer is yes. It's I
41:18
I wouldn't say yes answer is yes. It's I wouldn't say accurate answer depends on
41:20
wouldn't say accurate answer depends on
41:20
wouldn't say accurate answer depends on the topic and the question you're asking
41:23
the topic and the question you're asking
41:23
the topic and the question you're asking or work you're asking it to do but it
41:25
or work you're asking it to do but it
41:25
or work you're asking it to do but it has more features. For example, in chat
41:28
has more features. For example, in chat
41:28
has more features. For example, in chat GPT pro you can create these tasks. It
41:31
GPT pro you can create these tasks. It
41:31
GPT pro you can create these tasks. It can do things for you like agent can act
41:34
can do things for you like agent can act
41:34
can do things for you like agent can act an agent. So I can go to chat GPT pro
41:37
an agent. So I can go to chat GPT pro
41:37
an agent. So I can go to chat GPT pro and I can create a task saying hey chat
41:41
and I can create a task saying hey chat
41:41
and I can create a task saying hey chat GPT every morning 9:00 a.m. go do the
41:44
GPT every morning 9:00 a.m. go do the
41:44
GPT every morning 9:00 a.m. go do the research and give me all the latest news
41:46
research and give me all the latest news
41:46
research and give me all the latest news on generative AI and send it to my this
41:50
on generative AI and send it to my this
41:50
on generative AI and send it to my this email. So I have already done that. Let
41:51
email. So I have already done that. Let
41:51
email. So I have already done that. Let me see if I can show you
41:54
me see if I can show you
41:54
me see if I can show you um share my screen again.
41:57
um share my screen again.
41:57
um share my screen again. Let me share my screen quick. Okay, it
42:00
Let me share my screen quick. Okay, it
42:00
Let me share my screen quick. Okay, it may take more time to share and show
42:01
may take more time to share and show
42:01
may take more time to share and show you. But yeah, in uh pro mode there is
42:05
you. But yeah, in uh pro mode there is
42:05
you. But yeah, in uh pro mode there is something called I can go to this uh
42:08
something called I can go to this uh
42:08
something called I can go to this uh tasks and I can create a task I did
42:10
tasks and I can create a task I did
42:10
tasks and I can create a task I did already. By the way, what I did is I
42:12
already. By the way, what I did is I
42:12
already. By the way, what I did is I said every morning 9:00 a.m. find all
42:16
said every morning 9:00 a.m. find all
42:16
said every morning 9:00 a.m. find all the AI news, all generative AI news also
42:19
the AI news, all generative AI news also
42:19
the AI news, all generative AI news also news from this company and curate all
42:22
news from this company and curate all
42:22
news from this company and curate all that and send this to my email. Now
42:25
that and send this to my email. Now
42:25
that and send this to my email. Now every day 9:00 a.m. I automatically in
42:27
every day 9:00 a.m. I automatically in
42:27
every day 9:00 a.m. I automatically in my email I get all those AI news. Okay.
42:31
my email I get all those AI news. Okay.
42:31
my email I get all those AI news. Okay. I can also send I can connect all this
42:34
I can also send I can connect all this
42:34
I can also send I can connect all this to my other you can even connect to your
42:36
to my other you can even connect to your
42:36
to my other you can even connect to your I connected that to my email. Slowly
42:39
I connected that to my email. Slowly
42:39
I connected that to my email. Slowly what happens is these tools become more
42:41
what happens is these tools become more
42:41
what happens is these tools become more advanced and they will say oh and you
42:45
advanced and they will say oh and you
42:45
advanced and they will say oh and you can do this using other tools. Chad GP
42:47
can do this using other tools. Chad GP
42:47
can do this using other tools. Chad GP doesn't do right now but chat GPT now
42:49
doesn't do right now but chat GPT now
42:49
doesn't do right now but chat GPT now also can connect you with your outlook.
42:51
also can connect you with your outlook.
42:52
also can connect you with your outlook. It can say you can say create a task
42:54
It can say you can say create a task
42:54
It can say you can say create a task that every morning read my no every
42:56
that every morning read my no every
42:56
that every morning read my no every evening read my emails find me all the
42:58
evening read my emails find me all the
42:58
evening read my emails find me all the emails that that are sales related and
43:02
emails that that are sales related and
43:02
emails that that are sales related and make a summary and send this to my teams
43:04
make a summary and send this to my teams
43:04
make a summary and send this to my teams and reply this or send this to this
43:07
and reply this or send this to this
43:07
and reply this or send this to this person who can look at them so you can
43:09
person who can look at them so you can
43:09
person who can look at them so you can do all those kind of stuff and that's
43:10
do all those kind of stuff and that's
43:10
do all those kind of stuff and that's where future is going the future is
43:13
where future is going the future is
43:13
where future is going the future is going there's going to be lot of AI
43:15
going there's going to be lot of AI
43:15
going there's going to be lot of AI agents developed for doing different
43:17
agents developed for doing different
43:17
agents developed for doing different tasks and you are we are all start going
43:20
tasks and you are we are all start going
43:20
tasks and you are we are all start going to start using those agents
43:22
to start using those agents
43:22
to start using those agents So when you think about jobs, future
43:25
So when you think about jobs, future
43:25
So when you think about jobs, future jobs, where the future jobs are,
43:27
jobs, where the future jobs are,
43:27
jobs, where the future jobs are, building those agents, maintaining those
43:29
building those agents, maintaining those
43:29
building those agents, maintaining those agents, creating the LLMs, creating the
43:31
agents, creating the LLMs, creating the
43:31
agents, creating the LLMs, creating the data, in future, what will also happen
43:34
data, in future, what will also happen
43:34
data, in future, what will also happen is there's going to be these SLM small
43:37
is there's going to be these SLM small
43:37
is there's going to be these SLM small language models or between SLM and LLM
43:41
language models or between SLM and LLM
43:41
language models or between SLM and LLM focused on different industry, focused
43:43
focused on different industry, focused
43:43
focused on different industry, focused on different company, f focus on
43:45
on different company, f focus on
43:45
on different company, f focus on different people, right? For example, on
43:48
different people, right? For example, on
43:48
different people, right? For example, on C# corner, we are building right now SLM
43:51
C# corner, we are building right now SLM
43:51
C# corner, we are building right now SLM for our user base only. So chat GP
43:54
for our user base only. So chat GP
43:54
for our user base only. So chat GP doesn't know about that, but our model
43:56
doesn't know about that, but our model
43:56
doesn't know about that, but our model knows about it. So if you go to C# corn
43:58
knows about it. So if you go to C# corn
43:58
knows about it. So if you go to C# corn and ask a question about, oh, how many
44:01
and ask a question about, oh, how many
44:01
and ask a question about, oh, how many users in this episode ask questions
44:04
users in this episode ask questions
44:04
users in this episode ask questions about chat GPT? Now we have that data
44:08
about chat GPT? Now we have that data
44:08
about chat GPT? Now we have that data that's not public data. We are not
44:10
that's not public data. We are not
44:10
that's not public data. We are not sharing this data with open AI. So same
44:13
sharing this data with open AI. So same
44:13
sharing this data with open AI. So same thing will happen is if you want to go
44:16
thing will happen is if you want to go
44:16
thing will happen is if you want to go to a bank and you know ask bank may
44:19
to a bank and you know ask bank may
44:19
to a bank and you know ask bank may build their own chat right chat app they
44:21
build their own chat right chat app they
44:21
build their own chat right chat app they will bring their chat app where it will
44:23
will bring their chat app where it will
44:23
will bring their chat app where it will say it will you know you say oh what is
44:27
say it will you know you say oh what is
44:27
say it will you know you say oh what is my balance obviously this you can build
44:30
my balance obviously this you can build
44:30
my balance obviously this you can build easily chat GPT that what's your balance
44:32
easily chat GPT that what's your balance
44:32
easily chat GPT that what's your balance using any API uh for the bot framework
44:36
using any API uh for the bot framework
44:36
using any API uh for the bot framework you can build your own bot chatbot but
44:38
you can build your own bot chatbot but
44:38
you can build your own bot chatbot but it's not AI you now in future bank can
44:41
it's not AI you now in future bank can
44:41
it's not AI you now in future bank can build its own language model and that
44:45
build its own language model and that
44:45
build its own language model and that language model only works for the bank
44:47
language model only works for the bank
44:48
language model only works for the bank data and answers to the bank customers
44:51
data and answers to the bank customers
44:51
data and answers to the bank customers through a chatbot right things like that
44:53
through a chatbot right things like that
44:53
through a chatbot right things like that so that's where the future is more
44:55
so that's where the future is more
44:55
so that's where the future is more coming so in future we will be talking
44:59
coming so in future we will be talking
44:59
coming so in future we will be talking to more these apps through either voice
45:02
to more these apps through either voice
45:02
to more these apps through either voice command through our phone through our
45:04
command through our phone through our
45:04
command through our phone through our devices rather than logging to a online
45:07
devices rather than logging to a online
45:07
devices rather than logging to a online website and looking at my balance and
45:10
website and looking at my balance and
45:10
website and looking at my balance and transferring fund in future
45:12
transferring fund in future
45:12
transferring fund in future I can talk to my phone app. Bank has a
45:14
I can talk to my phone app. Bank has a
45:14
I can talk to my phone app. Bank has a phone app or an agent where I'll tell my
45:17
phone app or an agent where I'll tell my
45:17
phone app or an agent where I'll tell my bank, hey, can you transfer 100 rupees
45:19
bank, hey, can you transfer 100 rupees
45:19
bank, hey, can you transfer 100 rupees or $100 to this person from my account
45:22
or $100 to this person from my account
45:22
or $100 to this person from my account and this is their phone number and it
45:24
and this is their phone number and it
45:24
and this is their phone number and it will do it right there. So that's where
45:26
will do it right there. So that's where
45:26
will do it right there. So that's where the future is going and that's why
45:29
the future is going and that's why
45:29
the future is going and that's why learning these ROM techniques are
45:31
learning these ROM techniques are
45:31
learning these ROM techniques are important because they will become part
45:34
important because they will become part
45:34
important because they will become part of your software when you are building
45:36
of your software when you are building
45:36
of your software when you are building these agents because if you don't
45:37
these agents because if you don't
45:37
these agents because if you don't understand this part it'll be harder for
45:40
understand this part it'll be harder for
45:40
understand this part it'll be harder for you to build and understand the users.
45:46
Okay. So next question uh Michael has a
45:49
Okay. So next question uh Michael has a
45:49
Okay. So next question uh Michael has a question. What are the best ways to
45:50
question. What are the best ways to
45:50
question. What are the best ways to structure prompts and get better
45:51
structure prompts and get better
45:52
structure prompts and get better response? So Michael, I have two more
45:53
response? So Michael, I have two more
45:53
response? So Michael, I have two more episodes on that where we will go into
45:56
episodes on that where we will go into
45:56
episodes on that where we will go into more depth. Next Saturday we will talk
45:59
more depth. Next Saturday we will talk
45:59
more depth. Next Saturday we will talk about that exactly what are the you know
46:02
about that exactly what are the you know
46:02
about that exactly what are the you know how to structure your prompts, right? Um
46:05
how to structure your prompts, right? Um
46:05
how to structure your prompts, right? Um in the those next two episodes we'll
46:07
in the those next two episodes we'll
46:07
in the those next two episodes we'll focus on the week after we'll focus on
46:10
focus on the week after we'll focus on
46:10
focus on the week after we'll focus on advanced prom techniques like how do you
46:12
advanced prom techniques like how do you
46:12
advanced prom techniques like how do you advance because simple questions are
46:14
advance because simple questions are
46:14
advance because simple questions are simple questions but there will be time
46:16
simple questions but there will be time
46:16
simple questions but there will be time where you have these complex questions
46:18
where you have these complex questions
46:18
where you have these complex questions for business related and that's when you
46:20
for business related and that's when you
46:20
for business related and that's when you will need these advanced pro techniques.
46:25
Okay guys we have few more minutes. Is
46:27
Okay guys we have few more minutes. Is
46:27
Okay guys we have few more minutes. Is there any more questions? Great job
46:29
there any more questions? Great job
46:29
there any more questions? Great job everybody. Looks like a lot of people
46:31
everybody. Looks like a lot of people
46:31
everybody. Looks like a lot of people are hanging out out with us and watching
46:33
are hanging out out with us and watching
46:33
are hanging out out with us and watching the show. This is great. Any more
46:35
the show. This is great. Any more
46:35
the show. This is great. Any more questions Deepak we have or
46:40
um Thank you. Appreciate it everybody.
46:42
um Thank you. Appreciate it everybody.
46:42
um Thank you. Appreciate it everybody. Thank you so much. If you don't have any
46:44
Thank you so much. If you don't have any
46:44
Thank you so much. If you don't have any questions, we can close it. But this was
46:46
questions, we can close it. But this was
46:46
questions, we can close it. But this was great. Um
46:49
great. Um
46:49
great. Um so shades of hope everyday asks um there
46:53
so shades of hope everyday asks um there
46:53
so shades of hope everyday asks um there are different AIs and all give different
46:56
are different AIs and all give different
46:56
are different AIs and all give different answers even with prompt engineering. So
46:58
answers even with prompt engineering. So
46:58
answers even with prompt engineering. So which one should we trust? That's a
47:01
which one should we trust? That's a
47:01
which one should we trust? That's a great question. Remember AI
47:04
great question. Remember AI
47:04
great question. Remember AI AI is as good as your data, right? So
47:08
AI is as good as your data, right? So
47:08
AI is as good as your data, right? So all these different companies made
47:10
all these different companies made
47:10
all these different companies made different LLMs. LLM is where all the
47:14
different LLMs. LLM is where all the
47:14
different LLMs. LLM is where all the data is, right? And how this LLMs, how
47:18
data is, right? And how this LLMs, how
47:18
data is, right? And how this LLMs, how are these LLMs created?
47:21
are these LLMs created?
47:21
are these LLMs created? And what happened when OpenAI and these
47:23
And what happened when OpenAI and these
47:23
And what happened when OpenAI and these companies they went they stole all the
47:25
companies they went they stole all the
47:25
companies they went they stole all the data. I wouldn't say stool they copied
47:28
data. I wouldn't say stool they copied
47:28
data. I wouldn't say stool they copied which is almost like stealing without
47:29
which is almost like stealing without
47:30
which is almost like stealing without asking me right they took my articles
47:32
asking me right they took my articles
47:32
asking me right they took my articles from C# corner they took all our
47:34
from C# corner they took all our
47:34
from C# corner they took all our articles content they copied all those
47:37
articles content they copied all those
47:37
articles content they copied all those articles they copied code samples in
47:39
articles they copied code samples in
47:40
articles they copied code samples in their database then they work that's
47:42
their database then they work that's
47:42
their database then they work that's when we are when I talk about LLMs you
47:44
when we are when I talk about LLMs you
47:44
when we are when I talk about LLMs you will understand more details how they
47:46
will understand more details how they
47:46
will understand more details how they build the LMS so because they took data
47:49
build the LMS so because they took data
47:49
build the LMS so because they took data from different places they put data in
47:51
from different places they put data in
47:51
from different places they put data in different formats their engineers were
47:54
different formats their engineers were
47:54
different formats their engineers were different um some have their own data
47:57
different um some have their own data
47:57
different um some have their own data and that's why the outputs are different
48:00
and that's why the outputs are different
48:00
and that's why the outputs are different you will not find same output even even
48:03
you will not find same output even even
48:03
you will not find same output even even same what other thing is that when LLM
48:07
same what other thing is that when LLM
48:07
same what other thing is that when LLM the way they work they create new
48:08
the way they work they create new
48:08
the way they work they create new content every time you ask a question
48:11
content every time you ask a question
48:11
content every time you ask a question they don't copy content they don't use
48:14
they don't copy content they don't use
48:14
they don't copy content they don't use existing content even though they have
48:17
existing content even though they have
48:17
existing content even though they have copied the data from different websites
48:19
copied the data from different websites
48:19
copied the data from different websites different books um but they don't copy
48:22
different books um but they don't copy
48:22
different books um but they don't copy the data as it is it's not a database
48:24
the data as it is it's not a database
48:24
the data as it is it's not a database it's not a search. So what they do is
48:27
it's not a search. So what they do is
48:27
it's not a search. So what they do is they look at all the context and all
48:30
they look at all the context and all
48:30
they look at all the context and all that and they LLM create this new
48:34
that and they LLM create this new
48:34
that and they LLM create this new content every time you ask a question
48:37
content every time you ask a question
48:37
content every time you ask a question and that's why cost is big deal here
48:40
and that's why cost is big deal here
48:40
and that's why cost is big deal here right open AI is paying hundreds of
48:42
right open AI is paying hundreds of
48:42
right open AI is paying hundreds of millions of dollars every month not
48:45
millions of dollars every month not
48:45
millions of dollars every month not every year hundreds of millions of
48:46
every year hundreds of millions of
48:46
every year hundreds of millions of dollars every month to just run these uh
48:50
dollars every month to just run these uh
48:50
dollars every month to just run these uh LLM and that's like a billions of data
48:53
LLM and that's like a billions of data
48:53
LLM and that's like a billions of data billions of dollars every
48:55
billions of dollars every
48:55
billions of dollars every So they need to make all this money back
48:57
So they need to make all this money back
48:57
So they need to make all this money back and that's why their you know responses
49:00
and that's why their you know responses
49:00
and that's why their you know responses are different. It's like it's like
49:03
are different. It's like it's like
49:03
are different. It's like it's like imagine you have four people and you ask
49:05
imagine you have four people and you ask
49:05
imagine you have four people and you ask them same question you will get
49:07
them same question you will get
49:08
them same question you will get different answers. It's no different
49:10
different answers. It's no different
49:10
different answers. It's no different than that because they know more
49:11
than that because they know more
49:11
than that because they know more different knowledge right? If you ask
49:13
different knowledge right? If you ask
49:13
different knowledge right? If you ask somebody about you know what is Taj
49:16
somebody about you know what is Taj
49:16
somebody about you know what is Taj Mahel in UP they will have a different I
49:20
Mahel in UP they will have a different I
49:20
Mahel in UP they will have a different I mean not that's simple question answer
49:22
mean not that's simple question answer
49:22
mean not that's simple question answer is same it's a building blah blah but if
49:24
is same it's a building blah blah but if
49:24
is same it's a building blah blah but if you ask more questions everybody will
49:27
you ask more questions everybody will
49:27
you ask more questions everybody will answer different because based on what
49:29
answer different because based on what
49:29
answer different because based on what data they know what knowledge they have
49:31
data they know what knowledge they have
49:32
data they know what knowledge they have as simple as that
49:35
as simple as that
49:35
as simple as that okay guys this was great um everybody
49:37
okay guys this was great um everybody
49:37
okay guys this was great um everybody thank you for hanging out with me this
49:39
thank you for hanging out with me this
49:39
thank you for hanging out with me this Saturday I hope you learned something
49:41
Saturday I hope you learned something
49:41
Saturday I hope you learned something today I hope you learned learn a little
49:43
today I hope you learned learn a little
49:43
today I hope you learned learn a little bit more than what you knew yesterday.
49:45
bit more than what you knew yesterday.
49:45
bit more than what you knew yesterday. Um so yeah, if you want to continue to
49:48
Um so yeah, if you want to continue to
49:48
Um so yeah, if you want to continue to learn more about prompt engineering,
49:50
learn more about prompt engineering,
49:50
learn more about prompt engineering, LLMs, all these kind of tools, u just
49:55
LLMs, all these kind of tools, u just
49:55
LLMs, all these kind of tools, u just join us, continue watching us on C#
49:58
join us, continue watching us on C#
49:58
join us, continue watching us on C# corner. Every week we are coming. So
50:00
corner. Every week we are coming. So
50:00
corner. Every week we are coming. So next week we will focus on uh another
50:04
next week we will focus on uh another
50:04
next week we will focus on uh another prompting framework and then following
50:05
prompting framework and then following
50:06
prompting framework and then following week how to build learn advanced
50:07
week how to build learn advanced
50:07
week how to build learn advanced prompting new techniques. Follow us, let
50:10
prompting new techniques. Follow us, let
50:10
prompting new techniques. Follow us, let your friends know if you like it. And
50:12
your friends know if you like it. And
50:12
your friends know if you like it. And also don't forget to join and check out
50:14
also don't forget to join and check out
50:14
also don't forget to join and check out C# Corner where we are going to
50:16
C# Corner where we are going to
50:16
C# Corner where we are going to integrate this uh tokenization very soon
50:19
integrate this uh tokenization very soon
50:19
integrate this uh tokenization very soon and start learning these tokens. I will
50:21
and start learning these tokens. I will
50:21
and start learning these tokens. I will see you next week, same time.
50:27
[Music]
#Programming
#Education