πThis 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 β so β it if you find it valuable!
checkout - https://aiml-sheet.vercel.app/
| S.No |
Topic |
Description |
Resources |
| 0 |
Math for ML/DL |
Linear Algebra, Probability, Statistics, Calculus |
3Blue1Brown Β· CampusX |
| S.No |
Topic |
Description |
Resources |
| 1 |
Python Fundamentals |
Basics, data structures, file handling, exception handling, OOP |
FreeCodeCamp |
| S.No |
Topic |
Description |
Resources |
| 2 |
Streamlit Basics |
UI building, web apps for ML |
Chai aur Code |
4. Machine Learning β Core Basics
| S.No |
Topic |
Description |
Resources |
| 4 |
ML Fundamentals |
Classification, Regression, Pipelines, Feature Engineering |
CampusX Β· 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 Β· Roboflow |
| 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 β 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 |
| S.No |
Topic |
Description |
Resources |
| 12 |
Traditional NLP |
Text preprocessing, One-Hot Encoding, Bag of Words, TF-IDF, Word2Vec |
Krish Naik |
| 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 |
| S.No |
Topic |
Description |
Resources |
| 13 |
Deep Learning Fundamentals |
Neural Networks, Loss Functions, Optimizers, Activation Functions |
3Blue1Brown Β· MIT 6.S191 Campus X |
π 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 |
π 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 Β· 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
π 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 Β· LangChain RAG |
| 35 |
Advanced RAG |
Multi-query retrieval, re-ranking, hybrid search |
Pinecone RAG Guide |
| Project |
Description |
Tech Stack |
| PDF Q&A with RAG |
Upload PDF β extract β chunk β embed β query via LLM |
LangChain, FAISS, OpenAI/Groq, Streamlit |
| Multi-Document RAG |
Query across multiple documents with source attribution |
ChromaDB, LangChain, sentence-transformers |
| S.No |
Topic |
Description |
Resources |
| 36 |
Vector DB Fundamentals |
FAISS, ChromaDB, Pinecone, Weaviate, similarity search |
Pinecone Docs Β· ChromaDB |
| 37 |
Embedding Models |
sentence-transformers, OpenAI embeddings, custom embeddings |
Sentence Transformers |
| 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 Β· CrewAI |
| 41 |
ReAct & Planning |
Reasoning + Acting, chain-of-thought for agents |
ReAct Paper |
π 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 Β· LoRA Paper |
| 27 |
LoRA & QLoRA |
Low-Rank Adaptation, Quantized LoRA for efficient fine-tuning |
QLoRA Paper Β· Practical LoRA |
| 28 |
Quantization Techniques |
INT8, INT4, GPTQ, AWQ, GGML/GGUF formats |
BitsAndBytes Β· 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 Β· 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 |
π 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 Β· MCP GitHub |
| 46 |
MCP Implementation |
Building MCP servers, client integration, tool development |
Krishnaik |
20. FastAPI (Backend for AI)
π₯ Popular YouTube Channels
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
β Star this repository if you find it helpful!
This roadmap is continuously updated with the latest developments in Generative AI and Machine Learning.