AI, ML, Deep Learning & LLMs — Content & Cookbooks

A curated, hands-on library of notebooks, demos, and resources for AI/ML, Deep Learning, Generative AI, RAG, agents, fine-tuning, and modern tooling.

• PyPI: https://pypi.org/project/instructvault/
• GitHub: https://github.com/05satyam/instruct_vault

pip install instructvault

• PyPI: https://pypi.org/project/nonconvexoptimzationfunclib/
• GitHub: https://github.com/05satyam/nonconvexoptimizationfunclib

A Python library of standard non-convex benchmark functions for evaluating and comparing optimization algorithms.

pip install nonconvexoptimzationfunclib

1. Mission & Scope
2. Pick your path
3. Foundations (AI/ML Core)
4. LangChain (All Notebooks)
5. LlamaIndex (All Notebooks)
6. RAG Systems (All Variants)
7. Agents & Orchestration
8. Graph & Multimodal
9. MCP (Model Context Protocol)
10. Observability
11. Interview Experiences
12. Useful AI Agent Skills
13. External Resources
14. Utils
15. Repository Rules
16. Contributing & Support
17. License & Citation

This repository is a living library of practical AI/ML and Generative AI knowledge.
The focus is on learning by doing — notebooks and guides are reproducible, intuitive, and easy to extend.

If you're new, start with the path that matches your goal such as:

flowchart LR
  %% ========== Styles ==========
  classDef start fill:#E8F5E9,stroke:#2E7D32,stroke-width:1px,color:#1B5E20;
  classDef step  fill:#E3F2FD,stroke:#1565C0,stroke-width:1px,color:#0D47A1;
  classDef choice fill:#FFF8E1,stroke:#F57F17,stroke-width:1px,color:#E65100;
  classDef path  fill:#F3E5F5,stroke:#6A1B9A,stroke-width:1px,color:#4A148C;
  classDef build fill:#FFEBEE,stroke:#C62828,stroke-width:1px,color:#B71C1C;

  %% ========== Flow ==========
  A["🚀 Start Here<br/>Clone repo + setup venv"]:::start
  C{"Pick your learning path"}:::choice

  subgraph P["✨ Learning Paths"]
    direction TB
    F["🧠 Foundations<br/>Tokens • Embeddings • Transformers"]:::path
    H["🛠️ Hands-on Notebooks<br/>Run examples first"]:::path
    R["🔎 RAG Systems<br/>Naive → Hybrid → Graph → Multimodal"]:::path
    AG["🤖 Agents<br/>ReAct • Plan&Execute • Reflexion"]:::path
    I["🎯 Interview Prep<br/>System design + Q&A"]:::path
  end

  D["🔍 Observability & Evaluation<br/>Tracing • quality checks"]:::step
  E["🏗️ Build Mini-Projects<br/>End-to-end demo apps"]:::build
  Z["🌟 Keep iterating<br/>Add your own examples + PRs"]:::start

  %% ========== Edges ==========
  A --> C
  C --> F
  C --> H
  C --> R
  C --> AG
  C --> I

  F --> H
  H --> AG
  R --> AG
  AG --> D --> E --> Z

Everything that uses LangChain / LangGraph / LCEL lives here.

Everything that uses LlamaIndex lives here.

All Retrieval-Augmented Generation notebooks, grouped by type.

Agent skills are reusable SKILL.md playbooks that help coding agents follow better workflows for documents, coding, research, data work, planning, and tool usage. Start with trusted catalogs, review skills before installing them, and prefer project-local skills when you want repo-specific behavior.

• LLM Visualization
• Chatbot Arena LLM Leaderboard

• DeepLearning.AI — The Batch
• Uber AI — LLM Training
• Netflix ML Recommendations
• Sebastian Raschka

• Deep Dive into LLMs — Andrej Karpathy
• Deep Dive into LLMs like ChatGPT
• Software Is Changing (Again)
• Making AI accessible
• What is Agentic AI?
• Discover AI — Code4AI
• The AI GRID
• Krish Naik

• AI by Hand — Dr. Tom Yeh
• LLM Watch — Pascal Biese
• LangChain Blog
• LlamaIndex Blog
• MLOps Architect Mindset
• ByteByteGo Newsletter

• script_to_update_packages.py
• Steps2CreateEnvFile.MD
• requirements.txt

• Contributing Guide
• Code of Conduct
• Security Policy
• MIT License
• Citation File

Contributions are welcome!
If you spot an error, want a new notebook, or have an improvement idea:

• Read the Contributing Guide
• Open a PR / issue with a clear description

Security issues should be reported privately (see SECURITY.md).

This project is licensed under the MIT License — see the LICENSE file for details.

These notebooks reflect personal learnings and experiments.
Mistakes are part of the journey — use this repo as a starting point and adapt freely.

If this helps you, consider giving it a ⭐ on GitHub — it helps others find it too.