AI Handbook — From Zero to Hero

Table of Contents

1. Large Language Models

What are LLMs?

Large Language Models are AI systems trained on massive text data to understand and generate human-like text. Examples: GPT-4, Gemini, Claude, Llama.

How LLMs Work

• Tokenization: Breaking text into tokens (words/subwords)
• Embeddings: Converting tokens to numerical vectors
• Attention: Learning relationships between tokens
• Generation: Predicting next token based on context

Key Concepts

• Temperature: Controls randomness (0 = deterministic, 1 = creative)
• Tokens: Units of text (1 token ≈ 0.75 words)
• Context Window: Maximum tokens the model can process
• Fine-tuning: Customizing model for specific tasks

2. Prompt Engineering

Zero-Shot Prompting

Classify the following text as positive, negative, or neutral:
"The food was great and the service was excellent."
Answer: positive

Few-Shot Prompting

Classify the text as positive, negative, or neutral:
Text: "The food was great" → Positive
Text: "The service was terrible" → Negative
Text: "It was okay" → Neutral
Text: "I loved it" → 

Chain-of-Thought Prompting

Solve this step by step:
A store has 100 apples. They sell 30% on Monday and 20% of the remaining on Tuesday. How many apples are left?

Step 1: Monday sales = 100 * 0.30 = 30 apples
Step 2: Remaining after Monday = 100 - 30 = 70 apples
Step 3: Tuesday sales = 70 * 0.20 = 14 apples
Step 4: Remaining after Tuesday = 70 - 14 = 56 apples

Answer: 56 apples

3. API Development

import openai

def chat_with_gpt(prompt):
    response = openai.ChatCompletion.create(
        model="gpt-4",
        messages=[
            {"role": "system", "content": "You are a helpful assistant."},
            {"role": "user", "content": prompt}
        ],
        temperature=0.7
    )
    return response.choices[0].message.content

# Usage
answer = chat_with_gpt("What is data science?")
print(answer)

4. RAG Systems

What is RAG?

Retrieval-Augmented Generation (RAG) combines retrieval (searching a knowledge base) with generation (LLM) to provide accurate, up-to-date answers.

RAG Pipeline

1. Document Loading: Load PDFs, websites, databases
2. Text Splitting: Split documents into chunks
3. Embedding: Convert text to vectors
4. Vector Store: Store vectors in database
5. Retrieval: Search for relevant chunks
6. Generation: Generate answer using retrieved context

from langchain.document_loaders import PyPDFLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.embeddings import OpenAIEmbeddings
from langchain.vectorstores import FAISS

# Load document
loader = PyPDFLoader("document.pdf")
documents = loader.load()

# Split text
splitter = RecursiveCharacterTextSplitter(chunk_size=1000)
chunks = splitter.split_documents(documents)

# Create embeddings and vector store
embeddings = OpenAIEmbeddings()
vectorstore = FAISS.from_documents(chunks, embeddings)

# Search
results = vectorstore.similarity_search("What is data science?", k=3)

5. AI Agents

What are AI Agents?

AI agents are autonomous systems that can plan, reason, and use tools to accomplish tasks.

Agent Components

• LLM: The "brain" of the agent
• Tools: Functions the agent can use
• Memory: Short-term and long-term storage
• Planning: Breaking tasks into steps

from langchain.agents import initialize_agent
from langchain.llms import OpenAI
from langchain.tools import Tool

# Define tools
tools = [
    Tool(
        name="Search",
        func=search_web,
        description="Search the web for information"
    ),
    Tool(
        name="Calculator",
        func=calculate,
        description="Perform mathematical calculations"
    )
]

# Initialize agent
llm = OpenAI(temperature=0)
agent = initialize_agent(tools, llm, agent="zero-shot-react-description")

# Run agent
result = agent.run("What is the population of India divided by 2?")

6. AI Projects for Resume

7. AI Interview Questions

Download AI Handbook PDF

Get the complete AI Handbook as a free PDF. Enter your email to receive the download link.

Start Learning with DSWallah

Join 300+ students who transformed their careers with DSWallah. IIT-certified mentor, 50+ real projects, placement support.