GenAI_Roadmap.md
July 30, 2026 · View on GitHub
Learn AI/ML interactively at AI-ML Companion - Guided walkthroughs, architecture decisions, hands-on challenges, and narrated overviews for every project.
Your go-to hub for end-to-end GenAI learning. ⭐ Star this repo to stay updated with the latest GenAI resources :)
🔗 Quick Links
Interactive Roadmap | GenAI Essential Terms | AI/ML/Agentic AI Interview Q&A | GenAI Interview Q&A | Agentic AI Interview Q&A | GenAI Usecases | n8n Automation | GenAI on Azure | GenAI on AWS | GenAI on VertexAI |
📦 Comprehensive Prep Resources
Each phase is split by week and builds progressively.
Tip: Expand or collapse each phase as you go using dropdowns (▸) to get all the relevant links.
Phase 1: Foundation (Week 1-2)
| Topics | Core Material | Additional Resources | Code |
|---|---|---|---|
| AI vs ML vs DL vs GenAI | Video | GenAI Glossary Essential Terms (PDF) What is ML - interactive | |
| GenAI Intro | GenAI in Nutshell, LLMs by 3b1b | What is Generative AI (guide) LLMs & GenAI - interactive LLM Blog | |
| Python Basics | Crash Course, ML Crash | GitHub - Awesome Python Python for ML track - interactive | |
| ML & DL Basics | ML 101, Neural Networks | Crash DL ML Algorithms track - interactive Deep Learning track - interactive |
Phase 2: NLP & Transformer Core (Week 3-4)
Phase 3: LLM Fundamentals (Week 5-6)
| Topics | Core Material | Additional Resources | Code |
|---|---|---|---|
| Multimodal LLMs Basics | YT - How do Multimodal AI models work? | Multimodal LLMs - interactive | |
| Mixture of Experts (MoE) | 1. YT - What is Mixture of Experts? 2. Blog - Visual Guide to MoE | Blog - Huggingface MoE Mixture of Experts - interactive | |
| Open Weight LLMs (HF & Ollama) | 1. YT - Huggingface LLMs Locally 2. YT - Ollama | 1. Open-Weight Models - interactive 2. HuggingFace in Practice - interactive | 1. GitHub-HF 2. GitHub-Ollama |
| AI Frameworks | 1. LangChain 2. LlamaIndex | 1. LangChain Basics - interactive 2. Choosing a Framework - interactive | |
| Prompt Engineering | Coursera Specialization | Kaggle Whitepaper Prompt Engineering - interactive Advanced Prompting - interactive | GitHub |
Phase 4: RAG & Agentic Systems (Week 7-8)
Phase 5: Optimization & Evaluation (Week 9-10)
| Topics | Core Material | Additional Resources | Code |
|---|---|---|---|
| Fine Tuning | YT - Shaw Talebi | 1. Fine-Tuning - interactive 2. RLHF - interactive | 1. GitHub-ShawhinT 2. GitHub-Unsloth |
| Inference Optimization & Quantization | 1. Course - Quantization in Depth 2. Blog - GGUF 3. Blog - GGML | 1. Unsloth 2. GPTQModel 3. LLM Inference - interactive 4. Quantization & Compression - interactive | |
| Knowledge Distillation | Blog - Huggingface | Distillation - interactive | |
| LLM Evaluation | 1. Guide - LLM Evaluation 2. Tool - RAGA 3. Tool - Opik 4. Tool - DeepEval | 1. LLM Evaluation - interactive 2. Agent Evaluation - interactive 3. Blog |
Phase 6: Deployment & LLMOps (Week 11-12)
| Topics | Core Material | Additional Resources | Code |
|---|---|---|---|
| LLMOps | 1. Coursera Specialization 2. Deeplearning AI Course | 1. MLOps track - interactive 2. MLOps Intro - interactive 3. Agent Observability - interactive | |
| GenAI in Production | YT - RAG App on AWS | 1. Cloud GenAI track - interactive 2. Deploying Agents - interactive | GitHub Code |
| Safety & Alignment | 1. LlamaGuard 2. Prompt Guard | 1. LLM Security - interactive 2. Hallucinations - interactive 3. Agent Guardrails - interactive 4. Constitutional AI | Work in Progress |
| LLM Leaderboard | 1. Chatbot Arena 2. Artificial Analysis AI 3. Aider |
🛠️ Tools | Frameworks
| Category | Link |
|---|---|
| IDE & Dev Playground | Colab NoteBook, Kaggle Notebook, VSCode, Claude Code |
| LLM Aggregators | litellm, Openrouter, Aisuite |
| Agentic AI Frameworks | CrewAI, Relevance AI, LangGraph, SuperAGI, SmolAgent, OpenAI Agent SDK, Google's ADK Learn interactively: CrewAI, LangGraph, OpenAI Agents SDK |
| AI Automation (Workflow + AI Integration) | n8n, Zapier, Make |
| Cloud Agentic Framework | Microsoft Agent, AWS Bedrock Agent |
| Model Context Protocol (MCP) | YT- MCP Clearly Explained YT- MCP Crash Course for Python Developers Official Claude Doc GitHub-Awsome MCP Servers MCP Protocol - interactive |
| Agent2Agent Protocol (A2A) | Video - Demo Specification GitHub A2A Protocol - interactive |
| AI Coding Assistants | Claude Code, Gemini CLI, Cursor, Bolt, Lovable, Replit, V0 |
| Learning Platform | AI-ML Companion - Interactive AI/ML learning platform with 22 tracks, 300+ modules, visualizations, quizzes, and hands-on coding (ML fundamentals → LLMs → MLOps) |
💡 GenAI Use Cases
| Usecase | Link |
|---|---|
| Example Code for Running API-Based LLM | Groq, Claude, OpenAI, Gemini, DeepSeek, Cohere |
| Example Code for Running LLMs Locally | HuggingFace, Ollama Local, Ollama at Colab, LMStudio |
| Run FastMCP based Websearch Remote MCP Server Locally | GitHub |
| Production-Grade GenAI on AWS: RAG with Remote MCP Server | YouTube GitHub |
| Llama 4 Scout – All-in-One App for Chat, OCR, RAG & Agentic AI with CrewAI | GitHub |
| Advanced Snapshot-Based PDF Parsing using Docling, Markitdown, Gemini, Llama 4, Claude, GPT-4 & more | YouTube GitHub |
| Optical Character Recognition Using Advanced Vision-Language Models | GitHub |
| Chat with Neo4j Graph DB | YouTube GitHub |
| LLM-Based Chatbot for SQL Data Interaction | GitHub |
| LLM-Driven Chatbot to Interact with CSV Data | YouTube GitHub |
| Cache-Augmented Generation (CAG) | GitHub |
| AI Chat Analyzer for ChatGPT & Claud | GitHub |
🔖 Key Downloads
| Category | Links |
|---|---|
| Prompt Engineering | GitHub |
| RAG | |
| Agentic AI | GitHub |
| Misc | Embedding Models, GenAI Project Lifecycle |
The PDFs above are point-in-time snapshots. The always-current versions live on AI-ML Companion: GenAI Interview Questions · Agentic AI Interview Questions · LLM Interview Questions · ML Interview Questions · AI Engineer Interview Questions
📄 Imp Research Papers
| Title | Links |
|---|---|
| Kimi K2: Open Agentic Intelligence | Link |
| Gemini 2.5 | Link |
| DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning | Link |
| DeepSeek-V3 Technical Report | Link |
| Mixtral of Experts | Link |
