🚀This repo is all you need for AI

This cheatsheet serves as a practical roadmap and resource guide for anyone looking to get into GenAI or Agentic AI.

I'm actively exploring more resources and refining this roadmap to make it more detailed and genuinely helpful — so ⭐ it if you find it valuable!

checkout - https://aiml-sheet.vercel.app/

---

📋 Table of Contents

• Math Foundations
• Python Basics
• Streamlit
• FastAPI
• Machine Learning — Core Basics
• Machine Learning — Deep Dive
• ML for NLP
• Deep Learning Basics
• Core Deep Learning
• DL Frameworks
• MLOps
• Transformers
• Introduction to Gen AI
• Large Language Models (LLMs) - Advanced
• Introduction to LangChain
• RAG (Retrieval Augmented Generation)
• Vector Databases
• Agentic AI
• LangGraph & Advanced Agents
• Model Context Protocol (MCP)
• FastAPI (Backend for AI)
• Resources

---

0. Math Foundations

S.No	Topic	Description	Resources
0	Math for ML/DL	Linear Algebra, Probability, Statistics, Calculus	3Blue1Brown · CampusX

---

1. Python Basics

S.No	Topic	Description	Resources
1	Python Fundamentals	Basics, data structures, file handling, exception handling, OOP	FreeCodeCamp

---

2. Streamlit

S.No	Topic	Description	Resources
2	Streamlit Basics	UI building, web apps for ML	Chai aur Code

---

3. FastAPI

S.No	Topic	Description	Resources
3	FastAPI Fundamentals	REST APIs, async programming, model deployment	FastAPI Docs · FastAPI Course FastAPI Course

---

4. Machine Learning — Core Basics

S.No	Topic	Description	Resources
4	ML Fundamentals	Classification, Regression, Pipelines, Feature Engineering	CampusX · Stanford CS229
5	ML Evaluation	Accuracy, Precision, Recall, Confusion Matrix, ROC-AUC	StatQuest
6	Feature Scaling	Normalization, Standardization, MinMax, Robust Scaling	Scikit-learn Docs
7	Data Labeling	Manual annotation, Label Studio, Roboflow	Label Studio · Roboflow

🛠 P1: Core ML Projects

Project	Description	Datasets	Tech Stack
ML Classification App	Build a classification app using sklearn + Streamlit	Iris, Titanic, MNIST	sklearn, Streamlit, pandas
Regression Price Predictor	Housing price prediction with feature engineering	Boston Housing, California Housing	scikit-learn, seaborn, matplotlib

---

5. Machine Learning — Deep Dive

S.No	Topic	Description	Resources
8	Unsupervised ML	Clustering (K-Means, DBSCAN, Hierarchical), Dimensionality Reduction (PCA, t-SNE, UMAP)	StatQuest
9	Ensemble Methods	Bagging, Boosting (XGBoost, LightGBM), Stacking	Krish Naik
10	Hyperparameter Tuning	GridSearchCV, RandomSearch, Optuna, Bayesian Optimization	Optuna Docs
11	Core ML Concepts	Bias-variance tradeoff, Underfitting/Overfitting, Regularization (L1/L2)	Andrew Ng ML

---

6. ML for NLP

S.No	Topic	Description	Resources
12	Traditional NLP	Text preprocessing, One-Hot Encoding, Bag of Words, TF-IDF, Word2Vec	Krish Naik

🛠 P2: NLP Projects

Project	Description	Datasets	Tech Stack
Text Classifier	Spam detection or sentiment analysis using BoW/TF-IDF	SMS Spam, IMDb Reviews	sklearn, NLTK, pandas
Word2Vec Explorer	Visualize similarity between words using Word2Vec	Google News Word2Vec	Gensim, matplotlib, seaborn

---

7. DL Basics

S.No	Topic	Description	Resources
13	Deep Learning Fundamentals	Neural Networks, Loss Functions, Optimizers, Activation Functions	3Blue1Brown · MIT 6.S191 Campus X

---

8. Core DL

S.No	Topic	Description	Resources
14	Neural Networks & ANN	Feedforward networks, backpropagation, gradient descent	MIT 6.S191 · 3Blue1Brown Campus X
15	CNN	Convolutional Neural Networks for computer vision	Campus X MIT 6.S191 · CS231n
16	RNN & LSTM	Sequential data modeling, time series	Campus X MIT 6.S191 · Colah's Blog

---

9. DL Frameworks

S.No	Topic	Description	Resources
17	PyTorch/TensorFlow	Tensors, model building, training loops	PyTorch Docs · TensorFlow Docs · PyTorch Tutorial

🛠 P3: Deep Learning Projects

Project	Description	Datasets	Tech Stack
Image Classifier	Build CNN to classify cats vs dogs	Dogs vs Cats (Kaggle)	TensorFlow/Keras, PyTorch
Sentiment with LSTM	Sentiment prediction using LSTM networks	IMDb, Twitter Sentiment	Keras, PyTorch, torchtext

---

10. MLOps

S.No	Topic	Description	Resources
18	MLOps Fundamentals	Model versioning, experiment tracking, CI/CD for ML, monitoring	MLOps Playlist · MLOps Best Practices
19	Model Deployment	Docker, cloud deployment, model serving, A/B testing	MLOps Playlist
20	Experiment Tracking	MLflow, Weights & Biases, model registry	MLflow Docs · Weights & Biases

---

11. Transformers

S.No	Topic	Description	Resources
21	Transformer Architecture	Self-attention, Multi-head attention, Positional Encoding, Encoder-Decoder	3Blue1Brown · Campus X
22	Tokenization	BPE, SentencePiece, GPT-2 tokenizer, Hugging Face tokenizers	Campus X Andrej Karpathy · Original Paper

---

🛠 P4: Chatgpt from scratch

Project	Description	Tech Stack
Build an LLM from scratch	Build an LLM using transformers arch	Goat

---

12. Introduction to Gen AI

S.No	Topic	Description	Resources
23	GenAI Fundamentals	AI vs ML vs DL vs GenAI, How GPT/LLMs are trained, LLM evolution	Fireship · Two Minute Papers
24	LLM Evaluation	BLEU, ROUGE, Perplexity, Human Evaluation, Benchmarks	Hugging Face Evaluation
25	Ethics & AI Safety	Hallucination, bias, responsible deployment, alignment	AI Safety Course

---

13. Introduction to LangChain

S.No	Topic	Description	Resources
31	LangChain Fundamentals	Components, Chains, Agents, Memory	LangChain Docs · LangChain Tutorial
32	LLM Integration	OpenAI, Ollama, Hugging Face, Groq integration	Ollama Setup · Groq API
33	Prompt Engineering	Zero-shot, few-shot, chain-of-thought, prompt optimization	OpenAI Cookbook · Prompt Engineering Guide

🛠 P5: LangChain Projects

Project	Description	Tech Stack
Chatbot with LangChain	Build intelligent chatbot using LangChain + LLM + Streamlit	LangChain, Streamlit, Ollama/OpenAI
Document Summarizer	Summarize PDF/Text documents with LLMs	LangChain, PyPDF, Hugging Face Transformers

---

14. RAG (Retrieval Augmented Generation)

S.No	Topic	Description	Resources
34	RAG Fundamentals	Retrieval pipeline, embedding models, vector similarity	RAG Tutorial · LangChain RAG
35	Advanced RAG	Multi-query retrieval, re-ranking, hybrid search	Pinecone RAG Guide

🛠 P6: RAG Projects

Project	Description	Tech Stack
PDF Q&A with RAG	Upload PDF → extract → chunk → embed → query via LLM	LangChain, FAISS, OpenAI/Groq, Streamlit
Multi-Document RAG	Query across multiple documents with source attribution	ChromaDB, LangChain, sentence-transformers

---

15. Vector Databases

S.No	Topic	Description	Resources
36	Vector DB Fundamentals	FAISS, ChromaDB, Pinecone, Weaviate, similarity search	Pinecone Docs · ChromaDB
37	Embedding Models	sentence-transformers, OpenAI embeddings, custom embeddings	Sentence Transformers

---

16. Agentic AI

S.No	Topic	Description	Resources
38	AI Agent Fundamentals	Agent architecture, planning, tool use, memory systems	Lilian Weng's Blog
39	Tool-Using Agents	Function calling, external APIs, code execution	OpenAI Function Calling
40	Multi-Agent Systems	Agent collaboration, communication protocols	AutoGen · CrewAI
41	ReAct & Planning	Reasoning + Acting, chain-of-thought for agents	ReAct Paper

🛠 P7: Agentic AI Projects

Project	Description	Tech Stack
Research Assistant Agent	AI agent that can search web, summarize, and synthesize information	LangChain, Tavily/SerpAPI, OpenAI
Code Review Agent	Agent that reviews code, suggests improvements, runs tests	GitHub API, LangChain, code execution tools

---

17. Large Language Models (LLMs) - Advanced

S.No	Topic	Description	Resources
26	PEFT (Parameter Efficient Fine-Tuning)	LoRA, QLoRA, AdaLoRA, Prefix Tuning, P-Tuning	Hugging Face PEFT · LoRA Paper
27	LoRA & QLoRA	Low-Rank Adaptation, Quantized LoRA for efficient fine-tuning	QLoRA Paper · Practical LoRA
28	Quantization Techniques	INT8, INT4, GPTQ, AWQ, GGML/GGUF formats	BitsAndBytes · GPTQ
29	Model Compression	Pruning, Distillation, Quantization-Aware Training	Neural Compression
30	Advanced Fine-tuning	Full fine-tuning vs PEFT, Instruction tuning, RLHF basics	Hugging Face Fine-tuning

---

18. LangGraph & Advanced Agents

S.No	Topic	Description	Resources
42	LangGraph Fundamentals	State machines, graph-based workflows for agents	Campus X · LangGraph Tutorial
43	Complex Agent Workflows	Multi-step reasoning, conditional flows, human-in-the-loop	Campus X
44	Agent Orchestration	Managing multiple agents, workflow optimization	Campus X

🛠 P8: LangGraph Projects

Project	Description	Tech Stack
Multi-Step Research Agent	Agent that plans research, gathers info, and creates reports	LangGraph, multiple LLMs, web search APIs
Customer Service Agent	Complex customer service with escalation and human handoff	LangGraph, FastAPI, database integration

---

19. Model Context Protocol (MCP)

S.No	Topic	Description	Resources
45	MCP Fundamentals	Protocol for connecting AI assistants to external data sources and tools	Campus X · MCP GitHub
46	MCP Implementation	Building MCP servers, client integration, tool development	Krishnaik

---

20. FastAPI (Backend for AI)

S.No	Topic	Description	Resources
47	AI Model Deployment	Serving ML/DL models, batch processing, monitoring	MLOps Best Practices · MLOps Playlist

---

📚 Resources

🎥 Popular YouTube Channels

• 3Blue1Brown - AIML indepth intuition
• CampusX - Indian ML education
• Krish Naik - Comprehensive ML/AI tutorials
• IBM Technology - Fast recap while interviews
• Codebasics - extras
• FreeCodeCamp - extras
• Andrej Karpathy
• Jeremy Howard
• 3Blue1Brown
• Serrano Academy
• Lex Fridman
• Hamel Husain
• Machine Learning Street Talk
• Jason Liu
• StatQuest with Josh Starmer
• Dave Ebbelaar

📖 Essential Books

• All Top AIML Books - collection in one place

📄 Key Research Papers

• Attention Is All You Need - Original Transformer paper
• BERT: Pre-training of Deep Bidirectional Transformers
• GPT-3: Language Models are Few-Shot Learners
• LoRA: Low-Rank Adaptation of Large Language Models
• QLoRA: Efficient Finetuning of Quantized LLMs
• ReAct: Synergizing Reasoning and Acting in Language Models

🤝 Contributing

Feel free to contribute to this roadmap by:

• Adding new resources and tutorials
• Suggesting improvements to the learning path
• Sharing your project experiences
• Reporting broken links or outdated content

---

⭐ Star this repository if you find it helpful!

This roadmap is continuously updated with the latest developments in Generative AI and Machine Learning.

📌 Additional

• Follow me on X: https://x.com/lochan_twt
• Buy me a coffee: https://buymeacoffee.com/srilochan7p