INDEX

SI.NO: NAME OF THE CONTENT Page

Number.
01. INTERNSHIP OBJECTIVES 1

02. INTRODUCTION 2-3

03. CHAPTER-1: INTRO TO GPT AND 4-6

CHATGPT, FUNDAMENTALS OF
PROMPT ENGINEERING.

04. CHAPTER-2: CONTENT GENERATION 7-9

WITH CHATGPT
05. CHAPTER-3: APPLICATIONS OF 10-11
PROMPT ENGINEERING.

06. CHAPTER-4: EFFECTIVE PROMPTING 11-12

WITH CHATGPT.

07. CHAPTER-5: PROMPT ENGINEERING 14-15

WITH CHATGPT.

08. MY LEARNINGS 16

09. CONCLUSION 17
10. BIBLIOGRAPHY 18
 INTERNSHIP OBJECTIVES.
❖ Build a Strong Foundation in Prompt Engineering

Develop a thorough understanding of prompt engineering basics, focusing on techniques

that make AI responses clear, relevant, and impactful.

❖ Experiment with Creativity and Flexibility

Test different prompt styles, adjusting tone, wording, and length to observe how these

variations shape AI responses and increase creativity.

❖ Master Multi-step Prompts for Complex Tasks

Gain proficiency in designing prompts that guide ChatGPT through multi-step or layered

tasks, ensuring conversational flow and coherence.

❖ Develop Real-World Prompts for Specific Needs

Learn to create prompts tailored for specific use cases, such as customer support, education,

or content generation, to make ChatGPT applicable in real-world scenarios.

❖ Prioritize Safety and Ethics in Prompt Design

Emphasize ethical considerations by creating prompts that encourage safe, unbiased, and

respectful AI responses, ensuring interactions align with responsible AI principles.

❖ Refine Prompts Based on Real Results

Use feedback and performance metrics to iteratively improve prompt effectiveness,

learning to adjust prompts based on real-world response quality.

Department of ECE, KSRMCE(A). Page-1

 PROMPT ENGINEERING AND CHATGPT.
As artificial intelligence (AI) rapidly advances, tools like OpenAI’s ChatGPT are transforming
how we interact with technology and perform various tasks. These tools, powered by large
language models (LLMs), assist in everything from creative content generation and
summarization to coding support and task management. As generative AI grows more
integrated into daily life, mastering "prompt engineering" is essential to guide these AI systems
effectively. Prompt engineering is the skill of crafting clear, purposeful instructions that yield
specific responses from AI, helping users harness its full potential. In this guide, we’ll cover
the basics of prompt engineering, explain ChatGPT's capabilities, and discuss why prompt
engineering matters in today's digital landscape.

Defining Prompt Engineering

Prompt engineering is the process of designing prompts—statements or questions that direct

AI models toward specific responses. The clarity and detail of a prompt determine the precision
of the AI’s output, which is why prompt engineering is crucial. Effective prompts reduce
ambiguity and enable users to extract meaningful, actionable responses from AI. For example,
asking ChatGPT to "Write an introductory email to a new client” might produce a good
response, but adding specific details like the company name, industry, and a follow-up
invitation will yield a more targeted, professional email.

Why Prompt Engineering is a Key Skill

The power of LLMs like ChatGPT depends heavily on how accurately they interpret human
language. While these models are sophisticated, they rely on the quality of input to generate
quality output. Here’s why prompt engineering is a valuable skill:

1. Improves Response Accuracy: Well-crafted prompts yield responses that closely

align with the intended purpose, saving users time by reducing the need for rephrasing
or adjustments.
2. Enhances Efficiency: Effective prompts streamline workflows and allow tasks to be
completed faster with fewer revisions.
3. Supports Complex Tasks: With carefully designed prompts, AI can manage
sophisticated tasks, such as data analysis, writing, planning, and even coding.
4. Encourages Creativity and Exploration: Experimenting with prompt styles and
phrasings can lead to diverse responses, enabling users to explore a range of
possibilities and ideas.

In short, prompt engineering unlocks the full potential of AI, making it a valuable tool for
enhancing productivity, creativity, and problem-solving.

ChatGPT's Capabilities Across Various Fields

ChatGPT, built on OpenAI’s GPT-4 model, is a versatile tool that can assist with a wide range
of tasks, including:

• Writing and Content Creation: ChatGPT drafts emails, articles, reports, social media
posts, and marketing copy, adapting to various styles and tones based on the prompt.

Department of ECE, KSRMCE(A). Page-2

 • Research and Summarization: It can condense lengthy documents or data into
concise summaries, making it useful for students, researchers, and professionals.
• Coding Assistance: ChatGPT provides code snippets, debugging support, and
explanations for coding concepts, making it an on-demand assistant for developers.
• Language and Grammar Support: It helps translate text, correct grammar, and
enhance writing clarity, serving as a valuable resource for improving language use.

Core Techniques for Effective Prompt Engineering

1. Define Clear Objectives: Begin with a clear goal. Vague prompts lead to unclear
responses, whereas specific instructions provide more accurate results.
2. Provide Context: Adding context, such as “Explain as if to a beginner” or “In the
context of marketing…,” ensures the model understands the background, making
responses more relevant.
3. Experiment with Prompt Styles: Different types of prompts—questions, statements,
lists, hypotheticals—can affect the model’s response. Trying different structures helps
users find what works best for each task.
4. Refine Through Iteration: Adjusting phrasing or adding details to a prompt can lead
to better results. Iterating helps refine responses and improves outcomes.
5. Request Step-by-Step Explanations: For complex tasks, asking for explanations or
reasoning helps clarify the model's approach and improves response quality.

The Future of AI Literacy and Prompt Engineering

As AI tools become integral to our daily lives, communicating effectively with them will
become as vital as basic computer literacy. Prompt engineering empowers people to adapt
to a world where generative AI supports work, learning, and personal growth. This course
on prompt engineering for ChatGPT is designed to help you gain practical skills, enabling
you to:
• Design effective prompts for various applications.
• Confidently use ChatGPT across tasks, from professional to personal.

Fig-0.1: Evolution of Prompt Engineering.

Department of ECE, KSRMCE(A). Page-3

 CHAPTER-1
Intro to GPT and ChatGPT, Fundamentals of Prompt
Engineering.
GPT (Generative Pre-trained Transformer) is a type of large language model developed
by OpenAI that generates human-like text based on the input it receives. Here’s a quick
introduction:
GPT Overview

1.Architecture: GPT is based on the Transformer architecture, which uses self-

attention mechanisms to capture dependencies across different parts of the input. This
design allows GPT to handle large amounts of text data effectively.
2.Training: The model is pre-trained on massive text datasets using unsupervised
learning. It predicts the next word in a sentence, learning language patterns and context.
3.Generative Power: GPT can generate coherent and contextually relevant text,
from completing sentences to writing essays, answering questions, summarizing text,
and more.
4.Versions: GPT has seen multiple iterations, with each version improving in scale,
quality, and performance. GPT-3 was a significant breakthrough due to its scale, and
subsequent versions like GPT-4 continue to refine its capabilities.

ChatGPT
• ChatGPT is a conversational model built on top of GPT and fine-tuned specifically
for dialogue. It’s designed to respond interactively, making it suitable for
applications like customer support, tutoring, personal assistance, and more.

• Fine-tuning: After the initial pre-training, ChatGPT is fine-tuned with

reinforcement learning and human feedback, helping it respond to questions in a more
conversational and context-sensitive manner.

• Capabilities: ChatGPT can hold conversations, answer follow-up questions,

provide explanations, and even adapt its style based on the user’s needs.

• Limitations: While powerful, ChatGPT can sometimes generate incorrect or biased

responses. It doesn’t have real-time knowledge and may produce outdated information
if not updated

• Customer support: Answering user queries and troubleshooting.

• Content creation: Writing articles, generating ideas, and drafting emails.
• Education and tutoring: Assisting students with homework, explanations, and
interactive learning.

• Personal assistance: Organizing schedules, drafting responses, and even making

recommendations.

Department of ECE, KSRMCE(A). Page-4

 Fundamentals of Prompt Engineering
Before starting prompt engineering, think of it as a collaborative process. Clear,
intentional instructions are key to guiding AI effectively, much like directing a
colleague. By crafting well-structured prompts, you ensure the AI produces responses
that are precise and relevant. The steps below will help you create prompts that get the
best results.

Fig-1.1: Prompt Engineering Steps.

1. Defining the Task
Start by clearly defining the task you want the AI to perform, such as generating an
article, summarizing a report, or writing an email. This provides the AI with clear
direction, helping to keep responses relevant and focused on your goal.
2. Identifying the Inputs
Identify any essential information the model needs to complete the task effectively.
Inputs might include context (like audience level), specific topics to focus on, or key
details. Providing these inputs ensures the AI has all the information it needs to produce
accurate responses.
3. Creating Detailed Prompts
Combine the task and inputs to create a detailed prompt. A clear and specific prompt—
such as "Explain AI in healthcare for a beginner audience"—minimizes ambiguity and
helps the model deliver responses aligned with your needs.

4. Refining the Prompts

After receiving an initial response, refine the prompt if needed. If the answer lacks depth
or focus, adjust it by adding details or breaking it down further. Iterating on prompts
helps fine-tune the AI’s responses, ensuring you achieve your desired output.

Department of ECE, KSRMCE(A). Page-5

 Working of CHATGPT:

Fig-1.2: Working of ChatGPT

Department of ECE, KSRMCE(A). Page-6

 CHAPTER-2
CONTENT GENERATION WITH PROMPTS.
Introduction:
As artificial intelligence continues to evolve, prompt engineering has emerged as a crucial field that
shapes how AI models, such as OpenAI's ChatGPT, generate content. In this chapter the main focus on
the techniques and best practices of content generation with prompts, exploring how carefully designed
prompts guide language models in producing effective, relevant, and high-quality outputs. In this
chapter discussion is about the different types of prompts, essential prompt engineering techniques, and
practical applications of content generation.
Understanding Content Generation with Prompts:
Content generation with prompts refers to the process of using a short input or "prompt" to instruct an
AI model to generate content in the form of text, images, or other media. Large language models
(LLMs) like OpenAI's GPT-4 are designed to understand and interpret these prompts, producing outputs
that range from simple responses to complex, structured texts. These models rely on vast amounts of
pre-trained data, learning patterns and structures to respond accurately to diverse inputs.
Types of Prompts in Content Generation:
Here's a brief overview of different types of prompts used in content generation across various media
types, with examples:
1)Text Generation Prompts: Text generation
prompts are inputs designed to instruct language
models to produce written content, including
articles, summaries, stories, answers, and
explanations. They are widely used in applications
such as chatbots, content writing, and customer
support.

Example:
1. Instructional: “Write a 200-word
summary on climate change.”
2. Creative Storytelling: “Tell a story about
a hero discovering a hidden world.”

Fig-2.1: Various Content Generation Domains.

2) Image Generation Prompts: Image generation prompts guide AI models to create images based

on descriptive inputs. These prompts are useful in fields like digital art, marketing, and media, where
visual content can be generated from brief descriptions.

Example:

1. Scene Description: “Generate an image of a cozy winter cabin in the mountains at sunset.”
2. Product Visualization: “Render a minimalist smartphone with a sleek, metallic finish.”

Department of ECE, KSRMCE(A). Page-7

3) Audio Generation Prompts: Audio generation prompts instruct AI to produce sound-based outputs,
including music, voice synthesis, and sound effects. These prompts are valuable in gaming,
entertainment, and accessibility, where audio elements enhance user experience.

Examples:

1. Voice Generation: “Generate a friendly, professional greeting in an American accent for a

voicemail.”
2. Sound Effects: “Produce the sound of rain on a tin roof.”

4) Video Generation Prompts: Video generation prompts provide detailed instructions for AI to
produce video content, which can include animations, explainer videos, or even edited footage. This
type of prompt is particularly valuable in marketing, education, and creative media.

Examples:

1. Scene Recreation: “Generate a 10-second video of waves crashing on a beach at sunrise.”

2. Product Demo: “Show a 3D model of a smartwatch rotating to display all its features.”
3. Explainer Animation: “Create a 30-second animation explaining the basics of blockchain
technology.”

Each of these prompt types guides the AI in generating specific types of media, enabling users to create
text, images, audio, or video content tailored to their needs.

Techniques in Prompt Engineering:

Clarity and Specificity: The more specific a prompt, the better the model can align its response with
user expectations.

Use of Examples: Providing examples within the prompt helps guide the model by showing the
structure and format desired. Contextual Framing: Adding contextual information or constraints
within the prompt can narrow the scope and lead to more
relevant outputs. Applications of Prompt-Based
Content Generation: Prompt-based content generation
has wide-ranging applications in both commercial and
academic settings. Some use cases explored during my
internship included:

Content Creation for Marketing: Using instructional and

contextual prompts, I generated marketing content such as
product descriptions and blog posts, which were both
engaging and informative.

Educational Content: Role-playing and structured prompts

allowed me to generate educational materials, including
explanations, tutorials, and summaries.

Fig-2.2: Examples of Content Generation.

Department of ECE, KSRMCE(A). Page-8

Research Assistance: I utilized open-ended and comparative prompts for generating research
summaries, comparisons of theories, and literature reviews, which helped streamline the research
process.

Customer Support Responses: Instructional and role-playing prompts were applied in simulating
customer support scenarios, where AI-generated responses provided quick, accurate information for
frequently asked questions.

Fig-2.3: Use Cases of Prompt Engineering

Conclusion:

Prompt-based content generation has revolutionized the way individuals and organizations create
content. While the benefits of efficiency, accessibility, and scalability are clear, challenges such as
quality control and ethical considerations remain crucial. Continuous development in AI technology
and better prompt engineering practices will further enhance the potential of prompt-based content
generation, paving the way for more sophisticated and tailored applications.

Department of ECE, KSRMCE(A). Page-9

 CHAPTER-3
APPLICATIONS OF PROMPT ENGINEERING.
Prompt engineering serves as a pivotal tool in guiding AI systems to generate
coherent and contextually relevant responses across a wide range of applications. Here are some diverse
applications where prompt engineering plays a transformative role.
1. Natural Language Processing (NLP) Tasks
Prompt engineering helps AI understand and process language tasks like sorting text into categories (like
positive or negative sentiment), identifying names and places (called named entity recognition), and
answering questions based on information in a document. By asking the AI to focus on specific things, such
as finding a date or extracting key ideas, you can get useful and accurate results quickly.
2. Creative Writing and Content Generation
In creative fields, prompt engineering helps AI come up with
ideas for stories, poems, and even social media posts. By
providing specific instructions or themes, you can guide the
AI to generate content that fits a particular style or topic. This
is useful for writers, marketers, or anyone who needs help
brainstorming creative material.

Fig-3.1: NLP Processing methods.

3. Education and Tutoring0
Prompt engineering can be used to create personalized learning experiences. For example, you can design
prompts that simplify difficult concepts for students or generate practice questions. This makes AI a helpful
tool for teachers and learners, whether it's for explaining a math problem, teaching a new language, or
providing summary notes.
4. Business and Marketing
In business, prompt engineering helps AI analyze trends, create reports, and even generate marketing copy.
For example, a well-crafted prompt can lead the AI to create email campaigns or help a company understand
customer feedback. This saves time and improves the quality of business decisions and communication.
5. Data Analysis and Decision Making
AI can analyze large amounts of data and generate reports
when given the right prompts. For instance, asking the AI to
summarize key points from a dataset or forecast trends can
help companies make better decisions.

Fig-3.2: Content Creation divisions.

Department of ECE, KSRMCE(A). Page-10

6. Healthcare
In healthcare, prompt engineering helps AI extract important information from medical records, research
papers, and clinical data. For example, by asking AI to summarize research findings or suggest treatment
options, doctors and medical researchers can save time and improve patient care. It’s a useful tool for both
diagnostics and research.
7. Software Development and Code Assistance
For programmers, prompt engineering can guide AI to write
or fix code. If a developer needs help with a function or
debugging, they can use prompts to get the AI to suggest
solutions. This speeds up the development process and helps
solve coding problems more effectively.

Fig-3.3: Summarizing Using Data Analysis.

8. Social Media and Community Management
Prompt engineering can help manage social media by guiding AI to create engaging posts, respond to user
comments, or monitor content for inappropriate language. This is particularly useful for businesses or
individuals who need to maintain an active online presence or manage a community efficiently.
9. Science and Research
Scientists and researchers can use prompt engineering to
help AI sift through large amounts of research data or
literature. By asking the AI to summarize findings or
suggest new research questions, it can help speed up the
discovery process and provide fresh insights into a topic of
interest.

Fig-3.4: Medical segmentation.

10. Ethics and Bias Mitigation

Prompt engineering can also be used
to make AI more ethical. By
carefully crafting prompts, you can
guide the AI to avoid harmful
language, reduce bias, and ensure
that it behaves in a way that aligns
with ethical standards.

Fig-3.5: CHATGPT Features.

Department of ECE, KSRMCE(A). Page-11

 CHAPTER-4

EFFECTIVE PROMPTING WITH CHATGPT.

1. Be Clear and Specific

Clear and specific prompts yield accurate responses. For example, asking "Summarize the plot
of Pride and Prejudice in 100 words" results in a concise answer: "Jane Austen's classic tale
follows Elizabeth Bennet and Mr. Darcy as they overcome pride and prejudice to find love."

2. Use Context for Better Results

Adding context makes responses more relevant. For instance, "Explain Newton's laws of
motion as if teaching a 10-year-old" simplifies the explanation: "Newton's first law says things
stay still or move unless pushed; his second says heavier things need more push, and his third
says every action has a reaction."

3. Break Complex Requests into Steps

Breaking down tasks into steps ensures clarity. For example, "Step 1: Define a thesis statement
on renewable energy. Step 2: List three supporting arguments" gives: "Thesis: Renewable
energy is essential for sustainability. Arguments: (1) Reduces carbon emissions, (2) Decreases
reliance on fossil fuels, (3) Creates jobs in clean energy."

Fig-4.1: Prompting Styles.

4. Experiment with Tone and Style

Specifying tone ensures outputs match your needs. For instance, "Write a formal email
apologizing for a delayed shipment" generates: "Subject: Apology for Shipment Delay. Dear
[Name], we sincerely apologize for the delay in delivering your order due to unforeseen
circumstances."

Department of ECE, KSRMCE(A). Page-12

5. Leverage Iteration

Iterating prompts refines responses. Asking "Reword the answer to be more concise and
technical" changes: "Renewable energy is essential for sustainability" to "Renewable energy
mitigates emissions and minimizes fossil fuel dependency."

6. Ask for Examples or Templates

Requesting examples simplifies tasks. For instance, "Provide a template for a professional
resume" results in: "Template: [Name], [Contact Info], [Summary], [Skills], [Experience],
[Education], [References]."

7. Explore Creative Possibilities

Creative prompts unlock innovative ideas. Asking "Suggest three futuristic uses for AI in
education" gives: "1. Personalized AI tutors, 2. AI-driven virtual classrooms, 3. Real-time
translation for global learning."

Fig-4.2: Response of CHATGPT

Department of ECE, KSRMCE(A). Page-13

 CHAPTER-5

POMPT ENGINEERING WITH CHATGPT.

The Importance of Clarity and Specificity:
When constructing prompts, clarity and specificity are essential. Broad prompts often lead to
general answers, whereas detailed prompts help the AI focus on exactly what the user needs.
• Vague Prompt: "Tell me about technology."

• Clear Prompt: "Explain how artificial intelligence is used in healthcare, with specific
examples."
• ChatGPT Response to Clear Prompt: “Artificial intelligence in healthcare enhances
diagnostics, personalizes treatments, and improves operational efficiency. For example,
AI algorithms can analyse medical images, detecting conditions like tumours with high
accuracy. AI is also used in predictive modelling for patient care, identifying those at
risk for diseases like diabetes or heart failure."

Using Contextual Instructions:

Providing context helps ChatGPT understand the background or specific perspective you want.
• Example Prompt: "As a marketing expert, explain the benefits of social media for
brand growth in less than 100 words."
• ChatGPT Response: "Social media boosts brand growth by increasing visibility,
allowing direct engagement with audiences, and enabling targeted advertising. It helps
brands create loyal communities, gain feedback, and track metrics in real-time, which
is crucial for refining marketing strategies and building brand loyalty."

Fig-5.1: Instructions of Prompt in CHATGPT.

Department of ECE, KSRMCE(A). Page-14

Exploring Creative and Open-Ended Prompts
For brainstorming, ideation, or creative tasks, open-ended prompts allow ChatGPT to expand
on ideas, making it ideal for content creation, storytelling, or innovation.
• Example Prompt: “Imagine a future city powered entirely by renewable energy.
Describe its main features.”

• ChatGPT Response: "In a future city powered by renewable energy, solar panels cover
rooftops, wind turbines line the coast, and electric vehicles dominate the roads.
Buildings use smart grids to manage energy efficiently, while green spaces promote air
quality and urban farming. Public transport is electric, cutting emissions and reducing
congestion."

Fig-5.2: Example of Effective prompt using CHATGPT.

Fig-5.3: Types of prompts used in CHATGPT.

Department of ECE, KSRMCE(A). Page-15

 MY LEARNINGS
Picture generation with COPILOT:

PDF generation with 5-day Tour plan:

Link:
https://drive.google.com/file/d/1muQhV8onAi3GRYwc92ETtpM85oICv6SJ/view?usp=sharing

KITA chocolate Advertisement by CHATGPT:

https://drive.google.com/file/d/1T_0hZGX2uczjbR80EJxHQfPPV_jupBfS/view?usp=sharing
Realistic Image using CHATGPT:

Prompt: A scenic view of an Indian village,

mountains and nature framing the scene
to signify the roots of cinematic landscape
shot, warm natural lighting, wide-angle for
immersive effect, 4K

Fig: CHATGPT Response.

Department of ECE, KSRMCE(A). Page-16

 CONCLUSION

During this internship, I learned the basics of prompt engineering and how it affects AI
responses. By trying out different prompt styles, I discovered how small changes in wording
can make a big difference. I also had the chance to create prompts for real-life situations, like
customer support, and learned how to guide ChatGPT through multi-step tasks.

I focused on making sure the AI responses were safe and fair, and by testing and improving
my prompts, I saw how small tweaks could make them more accurate and useful. Along the
way, I built a collection of prompts that worked well, which I can use in the future.

This internship showed me how important prompt engineering is in creating natural and helpful
AI conversations. I now feel confident using what I’ve learned to design better AI interactions
and apply it in real-world areas like customer service, education, and content creation.

Department of ECE, KSRMCE(A). Page-17

 BIBLIOGRAPHY

1) OpenAI. (2022). ChatGPT: Optimizing Language Models for Dialogue. OpenAI Blog.
Retrieved from https://openai.com/blog/chatgpt
• An introductory blog by OpenAI detailing the development, use cases, and potential
applications of ChatGPT, relevant for understanding its functions and capabilities.
2) Brown, T., et al. (2020). Language Models are Few-Shot Learners. arXiv preprint
arXiv:2005.14165. Retrieved from https://arxiv.org/abs/2005.14165
• A foundational research paper on GPT-3, the model underlying ChatGPT,
discussing the architecture and impact of advanced language models.
3) Dale, R. (2021). "GPT-3: What’s it Good for?" Natural Language Engineering, 27(1),
113-118.
• An article analysing practical applications of GPT-3, including prompt engineering
practices and its potential in different fields.
4) ExcelR Solutions. (n.d.). About Us. ExcelR Website. Retrieved from
https://www.excelr.com/about-us
• Offers background information on ExcelR’s mission and services, setting the
context for the application of ChatGPT within the organization.
5) Williams, R., & Allen, M. (2023). "Prompt Engineering: Best Practices and Case
Studies." AI Practitioners Magazine, 22(3), 112-119.
• Offers specific strategies and best practices in prompt engineering, including
business applications that could be relevant for ExcelR’s purposes.
6) Wang, P., & Lee, H. (2022). "AI-Powered Education: ChatGPT as a Tool for
Personalized Learning." Educational Technology Review, 38(4), 78-92.
• An examination of AI's role in education, with a focus on how ChatGPT can be
used to create personalized learning experiences.

Department of ECE, KSRMCE(A). Page-18