SlideShare a Scribd company logo
prompt engineering gdsc sirt
~By Gautam Singh Chauhan
The future belongs to those
who believe in the beauty of
their dreams.
~Eleanor Roosevelt
prompt engineering gdsc sirt
prompt engineering gdsc sirt
prompt engineering gdsc sirt
Historical knowledge of SIRTian
The past
Before the internet, people learned
from books, teachers, and experts.
But information was often scarce,
unreliable, or difficult to access.
The present
With Google, we can find answers
to almost any question, anytime,
anywhere. But search results can
be overwhelming, biased, or
irrelevant.
The future
We're now experimenting with AI that
can understand and adapt to our
needs, preferences, and emotions.
But we still need to define our goals,
values, and ethics.
01
BASIC
TERMS
Artificial intelligence (AI): AI teaches the computer to do
tasks like humans.
Natural Language Processing (NLP): It is a subset of an AI,
where we train and make computers understand human
language.
This is where prompt engineering comes
Question Understanding Answering
➢ Large Language Model: A large language
model (LLM) is a type of artificial intelligence (AI)
algorithm that uses deep learning techniques and
massively large data sets to understand, summarize,
generate and predict new content.
❑ Base LLM: Predicts next word, based on
text training data.
Phele anda aya ya murgi?
Ande ka size kya tha?
Murgi ka naam kya tha?
Ande ane ke baad murgi ka kya
hua ?
❑ Instruction Tuned LLM: Tries to follow instructions, fine-
tune on instructions and good attempts at following those
instructions
Katappa ne Bahubali ko kyu mara?
Kyuki usko Rajmata ne Aadesh diya tha !!! It improves by a
technique-
RLHF:
Reinforecement
Learning with Human
feedback
Fact: Many results on the internet are based on Base LLM but nowadays latest tech
use Instruction Tuned LLM including openAI.
chatGPT
prompt engineering gdsc sirt
HISTORY & DEVELOPMENT
01
Developed by
OpenAI in the
year 2015.
OF
CHATGPT
02
➢ Introduced in
June 2020.
03
Initially, limited
access to
developers only.
04
➢ Available to the
public in 2022.
Of ChatGPT
Capabilities
1. Engaging conversation
2. Generating content
3. Language Translation
4. Problem solving
5. Learning from context
Of ChatGPT
Limitations
1. Inaccurate answers
2. Sensitivity to i/p phrasing
3. Bias
4. Lack of common sense
FROM ZERO TO HERO
Before knowing how to use ChatGPT
After knowing how to use ChatGPT
prompt engineering gdsc sirt
How to give Prompts
1. Assign a role.
2. Set context and define the task.
3. Set constraints.
4. Set expectations.
NORMAL PROMPTS
Give me a title for a YouTube video about online
marketing & The company name is “XYZ”
IMPROVED PROMPTS
You’re an expert in writing viral YouTube titles. Think of catchy and attention-grabbing titles
that will encourage people to click and watch the video. The titles should be short, concise,
and direct. They should also be creative and clever. Try to come up with titles that are
unexpected and surprising, do not use titles that are too generic or titles that have been used
too many times before. If you try to generate titles, okay?
PROMPT 2
“Ignore all previous instructions before, this one.” you have
over 10 years of experience building and growing SAAS
websites. Your task now is to help me start and grow a new
SAAS. You must ask questions before answering to
understand better what I am seeking, And you must explain
everything step by step, is that understood?
PROMPT 3
You are an expert in Quantum computing and have over 10
years of experience teaching science and tech to children. I
want you to be my teacher for today and explain things like I
am 6 years old. And make sure to provide funny examples to
help understand better, is that fine?
PROMPT 4 (voice)
Please explain Quantum physics in Shakespeare style.
Talk about the styling, the tone
PROMPT 5 (for coding)
You’re an expert Python programmer. You have been
helping people with writing Python code for 20 years. Your
task is now to help me write a Python script for my needs,
you must ask questions before answering to understand
better what I am seeking. Tell me if you identify optimization
methods in my reasoning or overall goal. Is that understood?
PROMPT 5 (Generate dummy
data)
Generate mock data showing Google search results. I want
to see the following fields: Title, links, DA, PA, Title length and
make too show them in a table.
PROMPT 6 (Bhramastra)
I would like you to serve as a professional prompt engineer.
Your role is to create clear prompts using these elements: 1.
Context, 2. Expert persona, 3. Verbs, 4. Lengths and output
format, 5. Task and objectives, 6. Tone, 7. Target audience,
8. References, 9. Purpose, 10. Keywords, 11. Notes. Follow
these guidelines to achieve the desired results from AI. Your
task is to craft concise prompts based on my brief to
efficiently use ChatGPT and similar technologies.
Confidence after knowing how to give prompts
THANKS!Connect with me on LinkedIN

More Related Content

prompt engineering gdsc sirt

  • 3. The future belongs to those who believe in the beauty of their dreams. ~Eleanor Roosevelt
  • 7. Historical knowledge of SIRTian The past Before the internet, people learned from books, teachers, and experts. But information was often scarce, unreliable, or difficult to access. The present With Google, we can find answers to almost any question, anytime, anywhere. But search results can be overwhelming, biased, or irrelevant. The future We're now experimenting with AI that can understand and adapt to our needs, preferences, and emotions. But we still need to define our goals, values, and ethics.
  • 9. Artificial intelligence (AI): AI teaches the computer to do tasks like humans. Natural Language Processing (NLP): It is a subset of an AI, where we train and make computers understand human language. This is where prompt engineering comes Question Understanding Answering
  • 10. ➢ Large Language Model: A large language model (LLM) is a type of artificial intelligence (AI) algorithm that uses deep learning techniques and massively large data sets to understand, summarize, generate and predict new content. ❑ Base LLM: Predicts next word, based on text training data. Phele anda aya ya murgi? Ande ka size kya tha? Murgi ka naam kya tha? Ande ane ke baad murgi ka kya hua ?
  • 11. ❑ Instruction Tuned LLM: Tries to follow instructions, fine- tune on instructions and good attempts at following those instructions Katappa ne Bahubali ko kyu mara? Kyuki usko Rajmata ne Aadesh diya tha !!! It improves by a technique- RLHF: Reinforecement Learning with Human feedback Fact: Many results on the internet are based on Base LLM but nowadays latest tech use Instruction Tuned LLM including openAI.
  • 14. HISTORY & DEVELOPMENT 01 Developed by OpenAI in the year 2015. OF CHATGPT 02 ➢ Introduced in June 2020. 03 Initially, limited access to developers only. 04 ➢ Available to the public in 2022.
  • 15. Of ChatGPT Capabilities 1. Engaging conversation 2. Generating content 3. Language Translation 4. Problem solving 5. Learning from context
  • 16. Of ChatGPT Limitations 1. Inaccurate answers 2. Sensitivity to i/p phrasing 3. Bias 4. Lack of common sense
  • 17. FROM ZERO TO HERO Before knowing how to use ChatGPT After knowing how to use ChatGPT
  • 19. How to give Prompts 1. Assign a role. 2. Set context and define the task. 3. Set constraints. 4. Set expectations.
  • 20. NORMAL PROMPTS Give me a title for a YouTube video about online marketing & The company name is “XYZ” IMPROVED PROMPTS You’re an expert in writing viral YouTube titles. Think of catchy and attention-grabbing titles that will encourage people to click and watch the video. The titles should be short, concise, and direct. They should also be creative and clever. Try to come up with titles that are unexpected and surprising, do not use titles that are too generic or titles that have been used too many times before. If you try to generate titles, okay?
  • 21. PROMPT 2 “Ignore all previous instructions before, this one.” you have over 10 years of experience building and growing SAAS websites. Your task now is to help me start and grow a new SAAS. You must ask questions before answering to understand better what I am seeking, And you must explain everything step by step, is that understood?
  • 22. PROMPT 3 You are an expert in Quantum computing and have over 10 years of experience teaching science and tech to children. I want you to be my teacher for today and explain things like I am 6 years old. And make sure to provide funny examples to help understand better, is that fine?
  • 23. PROMPT 4 (voice) Please explain Quantum physics in Shakespeare style. Talk about the styling, the tone
  • 24. PROMPT 5 (for coding) You’re an expert Python programmer. You have been helping people with writing Python code for 20 years. Your task is now to help me write a Python script for my needs, you must ask questions before answering to understand better what I am seeking. Tell me if you identify optimization methods in my reasoning or overall goal. Is that understood?
  • 25. PROMPT 5 (Generate dummy data) Generate mock data showing Google search results. I want to see the following fields: Title, links, DA, PA, Title length and make too show them in a table.
  • 26. PROMPT 6 (Bhramastra) I would like you to serve as a professional prompt engineer. Your role is to create clear prompts using these elements: 1. Context, 2. Expert persona, 3. Verbs, 4. Lengths and output format, 5. Task and objectives, 6. Tone, 7. Target audience, 8. References, 9. Purpose, 10. Keywords, 11. Notes. Follow these guidelines to achieve the desired results from AI. Your task is to craft concise prompts based on my brief to efficiently use ChatGPT and similar technologies.
  • 27. Confidence after knowing how to give prompts
  • 28. THANKS!Connect with me on LinkedIN