Generative AI Tutorial
Generative AI is a branch of artificial intelligence that focuses on creating new content such as text, images, code, music and video using models like transformers, GANs and diffusion models. It is used in tools like ChatGPT, Claude, Gemini and DALL·E and is widely applied in automation, chatbots and personalization.
This Generative AI tutorial offers a step-by-step guide to all major concepts and techniques required to learn and build GenAI applications, with practical projects and modern frameworks.
1. Tools for Generative AI
To get started with Generative AI, you need to build expertise in the following tools and libraries:
- Python
- PyTorch
- TensorFlow
- Hugging Face Transformers
- LangChain
- LangGraph
- Langflow
- LlamaIndex
- Integration of Langchain with Llama-Index
2. Core Concepts in Generative AI
Understanding the foundations of AI and deep learning is essential for working with GenAI models.
- What is Artificial Intelligence?
- What is Generative AI?
- Neural Networks
- RNNs, LSTMs, GRUs
- Transformers and Self-Attention
- Autoencoders and Latent Space
- GANs and Diffusion Models
3. Natural Language Processing (NLP) Basics
Most Generative AI models are built on NLP concepts.
- Text Preprocessing in NLP
- Bag of Words & TF-IDF
- Word2Vec & GloVe
- Introduction to BERT
- Introduction to GPT Models
- Hugging Face Models
4. Prompt Engineering
Prompt engineering is the practice of crafting inputs to get better outputs from LLMs.
- What is Prompt Engineering?
- Zero-Shot, One-Shot and Few-Shot Prompting
- Chain of Thought Prompting
- Role & Contextual Prompting
- ReAct (Reasoning + Acting) Prompting
- Retrieval-Augmented Prompting
- Self-Consistency Prompting
- Tree of Thought (ToT) prompting
- Guardrails in AI
5. Large Language Models (LLMs)
LLMs are the backbone of modern Generative AI systems.
- Large Language Model
- LLM Parameters
- Scaling Laws in LLMs
- Fine-Tuning LLMs with LoRA , QLoRA and PEFT
- RLHF: Reinforcement Learning from Human Feedback
- LLM Distillation
- Popular LLMs: GPT, Claude, LLaMA, Gemini
- LLM APIs: OpenAI, Hugging Face, Gemini
6. Retrieval-Augmented Generation (RAG)
RAG combines LLMs with external knowledge sources for more accurate responses.
- RAG in AI
- RAG Architecture
- Multimodal RAG
- Embeddings
- Vector Databases: FAISS, ChromaDB, Qdrant, Pinecone
- RAG System with Langchain and Langraph
7. Agentic AI & Multi-Agent Systems
Agentic AI extends LLMs with autonomy, memory and collaboration.
- What is Agentic AI?
- Agent vs Traditional AI
- Agent Architectures & Memory
- Agent-to-Agent Communication
- AI Agent Frameworks
- Model Context Protocol (MCP)
8. CrewAI and Orchestration
CrewAI is a framework for coordinating multiple AI agents to work collaboratively.
- Introduction to CrewAI
- CrewAI Tools
- Creating Custom Tools for CrewAI
- Memory in CrewAI
- CrewAI Embeddings
- CrewAI Collaboration
- CrewAI Knowledge
- CrewAI Planning and Reasoning
- CrewAI CLI
- CrewAI Flow
- Fraud Detection Using CrewAI Project
9. Automation with Agents and Deployement
Generative AI can be extended into workflows for business automation.
- Agentic RAG
- Agentic RAG with LlamaIndex
- Introduction to n8n
- Automated Email Classifier with n8n
- AI Deployment with Gradio, Streamlit, FastAPI
10. Responsible & Ethical AI
Generative and Agentic AI raise ethical challenges that must be addressed.
11. Projects
Practical, hands-on projects are essential for mastering agentic AI. This section provides real-world project ideas to build your portfolio.
- Building an AI application with LlamaIndex
- PDF Summarizer LLM Application
- RAG(Retrieval-Augmented Generation) using LLama3
- Simple Retrieval Augmented Generation using Java
- Building AI Agents with Phidata
- Multimodal RAG using Python
- PDF Summarizer using RAG
12. Careers in Generative & Agentic AI
Generative AI and Agentic AI are among the fastest-growing career domains in tech. Key job roles include:
