🟩 PART A — Python Foundations

Module 1: Python Programming Foundations

● Python setup, syntax, variables, and data types

● Operators, conditional statements (if/elif/else), and loops (for/while)

● Functions, modules, and error handling in Python

Module 2: Python Data Structures & File Handling

● Lists, tuples, dictionaries, sets, and comprehensions

● String methods, indexing, and slicing

● Reading and writing files, including CSV and JSON formats

Module 3: Object-Oriented Python & APIs

● Classes, objects, inheritance, and polymorphism

● Exception handling and clean error management

● Web scraping with Beautiful Soup and Requests; working with APIs

Module 4: Data Analysis with Python

● Introduction to NumPy and Pandas for data manipulation

● Working with databases (SQLite)

● Data visualization with Matplotlib and Seaborn

✅ PART B — Machine Learning

Module 5: Introduction to Machine Learning

● Machine learning applications and the end-to-end ML workflow

● Supervised vs. unsupervised learning

● Setting up the ML environment — Jupyter, NumPy, Pandas, Scikit-learn

Module 6: Data Wrangling & Feature Engineering

● Handling missing values and outliers

● Categorical encoding — One-Hot and Label Encoding

● Feature scaling, feature selection, and dimensionality reduction

● Project: Analysis of the Titanic Dataset

Module 7: Regression Algorithms

● Simple and multiple linear regression; MSE, RMSE, R², regularization

● Logistic regression — sigmoid function, binary classification

● Evaluation metrics: accuracy, precision, recall, F1 score, confusion matrix, AUC

● Projects: House Price Prediction, Heart Disease Prediction

Module 8: Classification & Ensemble Algorithms

● Decision Trees and Random Forest

● k-Nearest Neighbours (KNN) and Support Vector Machines (SVM)

● Naive Bayes

● Project: Breast Cancer Prediction — comparing models to find the best fit

Module 9: Model Evaluation, Tuning & Clustering

● Cross-validation, train-test split, and the bias-variance trade-off

● Hyperparameter tuning with Grid Search and Random Search; handling imbalanced datasets

● K-Means, Hierarchical Clustering, DBSCAN, and dimensionality reduction with t-SNE

● Project: Customer Segmentation

🚀 PART C — Generative AI

Module 10: Foundations of Generative AI

● Evolution of NLP and neural language models

● Generative vs. discriminative models; types of generative AI models

● Python libraries for GenAI — Hugging Face Transformers, OpenAI; ethical considerations

Module 11: Transformers & Attention Mechanism

● Self-attention and multi-head attention

● Encoder-decoder architecture and positional encoding

● Pre-training/fine-tuning paradigm; modern architectures (GPT, BERT, T5)

Module 12: Prompt Engineering & LLM APIs

● Zero-shot, one-shot, few-shot, and chain-of-thought prompting

● Role prompting, temperature/sampling strategies, structured output generation

● API integration — authentication, streaming responses, rate limiting

● Project: Chatbot built with the OpenAI API

Module 13: Token & Cost Optimization

● Tokenization, token counting, and context window management

● Input chunking strategies

● Cost analysis and strategies for reducing API costs across LLM providers

Module 14: Retrieval-Augmented Generation (RAG)

● RAG architecture, document processing, and chunking

● Vector databases, embedding models, and similarity search

● Query reformulation, context augmentation, and hybrid search

● Project: Question-Answering System built with RAG

Module 15: Lang Chain Framework

● Chains, sequential processing, prompt templates, and output parsers

● Memory types for conversation management

● Tools, agents, document loaders, and text splitters

● Project: Multi-Tool Assistant built with LangChain

Module 16: Fine-Tuning, Quantization & LLM Evaluation

● Parameter-efficient fine-tuning — LoRA, QLoRA, adapter-based methods

● Model quantization and deployment considerations

● Evaluation metrics, benchmark datasets, and evaluating hallucinations

Module 17: Agentic AI Workflows

● Planning and reasoning in AI agents; the ReAct framework

● Tool and API usage by agents; multi-agent systems

● Memory, state management, and feedback loops

● Project: Task-Oriented Agent

Module 18: Capstone Project

● Problem definition and architecture design for a full data science / GenAI application

● Implementation, UI development (Streamlit/Gradio), and performance optimization

● Testing, evaluation, deployment, and documentation

Hands-on Projects

1. Titanic Dataset Analysis — Clean and explore a   real-world dataset, handle missing values, and visualize survival patterns   using Pandas and Matplotlib.

2.House Price Prediction — Build a linear   regression model to predict housing prices, evaluated with MSE, RMSE, and R².

3.Heart Disease Prediction — Build a logistic   regression classifier for binary medical diagnosis, evaluated with precision,   recall, and AUC.

4.Breast Cancer Prediction — Compare Decision   Trees, Random Forest, KNN, SVM, and Naive Bayes to find the best-performing   classifier.

5.Customer Segmentation — Apply K-Means clustering   to group customers by behaviour for targeted business strategy.

6. LLM Chatbot — Build a conversational chatbot   using the OpenAI API with prompt engineering and streaming responses. 

7.RAG Question-Answering System — Build a   retrieval-augmented generation pipeline using vector search and embeddings to   answer questions from a custom knowledge base.

8. Multi-Tool AI Assistant — Build a   LangChain-powered assistant that chains tools, memory, and external APIs into   one workflow.

9. Task-Oriented AI Agent (Capstone) — Design and   deploy an end-to-end agentic AI application with planning, tool use, and a   Streamlit/Gradio interface.