Machine Learning, Deep Learning, Generative AI, LLMs, RAG, AI Agents, MLOps, Projects, Deployment, Open Source & Career Guide
git clone https://github.com/udaysharmadev/Ai-Roadmap.git && cd Ai-Roadmap
pip install -r requirements.txt
pytest -q # 90 tests, zero API keys
python -m ai_roadmap.progress --summary # requires pip install -e . (or PYTHONPATH=src)Then pick a route: learn (notebooks/01_… → 05_… in order),
build (projects/01_… → 07_…), or test yourself (docs/quizzes/).
In 2026, learning Artificial Intelligence can feel overwhelming due to information overload. This repository is built as a practical, step-by-step, zero-to-advanced guide mapped with free, high-quality YouTube resources, production code templates, and project tiers. It is designed to help you become an industry-ready AI Engineer, ML Engineer, or Startup Founder.
- Students & Beginners looking to start from mathematical foundations.
- Software Developers wanting to transition into Generative AI and LLMs.
- Freelancers who want to offer custom RAG and Agentic solutions to clients.
- Startup Founders aiming to build AI SaaS applications.
Here are the specialized pathways available in this repository:
| Pathway | Focus | Documentation |
|---|---|---|
| 🤖 Machine Learning | Math foundations, NumPy, Pandas, Scikit-Learn | View Guide |
| 🧠 Deep Learning | Neural Networks, CNNs, RNNs, PyTorch, Transformers | View Guide |
| 👁️ Computer Vision | Image Processing, OpenCV, YOLO, Segmentation | View Guide |
| 💬 NLP | Tokenization, Word Embeddings, BERT, Sequence Tagging | View Guide |
| 🚀 Generative AI | Prompting, API Integration, LangChain, LlamaIndex | View Guide |
| 🗂️ RAG Systems | Chunking, Embeddings, Vector DBs, Hybrid Search | View Guide |
| 🤖 AI Agents | Autonomous loops, ReAct, CrewAI, LangGraph, MCP | View Guide |
| ⚙️ MLOps & Ops | Docker, FastAPI, Git, Streamlit, Cloud Hosting | View Guide |
| If you are… | Do this | Prove it |
|---|---|---|
| Student / Beginner | Phases 1→3, notebooks 01–03 |
Phase 1–3 quizzes, Milestones 1–3 |
| Developer → GenAI | Phases 4→5, notebooks 04–05, template streamlit_rag |
Phase 5 quiz, Milestone 5 |
| Freelancer | Projects 05–07, freelancing guide |
Deployed RAG demo URL + brief |
| Founder | Full loop + docker compose up --build |
Live API + chat URL (Milestone 7) |
| Job seeker | Interview prep + flashcards | 40/40 cards, mock Q&A |
graph TD
classDef purple fill:#a855f7,stroke:#a855f7,stroke-width:2px,color:#fff;
classDef blue fill:#3b82f6,stroke:#3b82f6,stroke-width:2px,color:#fff;
classDef emerald fill:#10b981,stroke:#10b981,stroke-width:2px,color:#fff;
classDef orange fill:#f97316,stroke:#f97316,stroke-width:2px,color:#fff;
classDef pink fill:#ec4899,stroke:#ec4899,stroke-width:2px,color:#fff;
classDef teal fill:#14b8a6,stroke:#14b8a6,stroke-width:2px,color:#fff;
Phase1[Phase 1: Math & Python Foundations]:::purple --> Phase2[Phase 2: NumPy, Pandas & Datasets]:::blue
Phase2 --> Phase3[Phase 3: Scikit-Learn & Machine Learning]:::emerald
Phase3 --> Phase4[Phase 4: Neural Networks & PyTorch]:::orange
Phase4 --> Phase5[Phase 5: CV, NLP, GenAI & Agents]:::pink
Phase5 --> Phase6[Phase 6: Multi-Tier AI Projects]:::pink
Phase6 --> Phase7[Phase 7: FastAPI, Docker & MLOps Deployment]:::teal
Learn the core languages and logical mathematics that power data transformations.
- 🐍 Python Full Course: CodeWithHarry Playlist
- 📐 Linear Algebra: CampusX Playlist
- 📊 Statistics for AI: CampusX Video Guide
- 🎲 Probability: CampusX Video Guide
- 💻 Jupyter & Colab Guide: Setup Tutorial
Master the essential tools for reading, cleaning, and visualizing structured datasets.
- 🧮 NumPy Tutorial: freeCodeCamp Guide
- 🐼 Pandas Full Course: CampusX Playlist
- 📈 Matplotlib & Seaborn Visualizations: CampusX Playlist
Understand regression, classification, clustering, and evaluating model metrics.
- 🤖 100 Days of Machine Learning: CampusX Course
- 📉 Feature Engineering & Imputation: CampusX Playlist
- 📊 StatQuest Machine Learning Videos: Concept Explanations
- 🎯 Evaluation Metrics Tutorial: Accuracy, Precision, Recall
Build deep neural networks, CNNs, sequence models, and attention mechanisms.
- 🧠 Neural Networks Zero to Hero: Andrej Karpathy Course
- 🔥 PyTorch Deep Learning Bootcamp: freeCodeCamp Guide
- 🛠️ TensorFlow Developer Course: freeCodeCamp Guide
- ⚡ Transformers from Scratch: Karpathy GPT Guide
Dive into modern state-of-the-art Generative AI architectures, RAG systems, and autonomous agents.
- 🚀 Generative AI for Developers: Andrew Ng Course Intro
- 💬 Intro to Large Language Models: Karpathy Guide
- 🤖 AI Agents & Tool Use: ReAct Agent Guide
- 🔌 Model Context Protocol (MCP): MCP Tutorial Guide
- 🗂️ Vector Databases & RAG from Scratch: Pinecone & LangChain
- 🧪 PEFT & QLoRA Fine-Tuning: Fine-tuning LLMs Guide
Build real-world systems following our step-by-step guides in docs/projects-roadmap.md.
- 📧 Spam Classifier: Tutorial Guide
- 🎥 Recommendation Engine: Tutorial Guide
- 💬 AI Customer Chatbot: Local Ollama Chatbot
- 📄 Resume Skill Analyzer: SpaCy NLP Parsing
- 🗂️ PDF Q&A Assistant: RAG Tutorial
- 🎙️ Voice AI Booking Assistant: Multi-Agent Voice Bot
- 🌐 AI Agent Team SaaS: CrewAI multi-agent group
Ship your models to production and make them available to users worldwide.
- 🐳 Docker Containers: Docker Crash Course
- ⚡ FastAPI APIs: FastAPI Tutorial
- 📊 Streamlit UI: Rapid Dashboard Prototyping
- ☁️ Deploying to Cloud PaaS: Render & Railway Deployment
- ⚙️ MLOps experiment tracking: MLflow Setup
If you want to tailor your learning towards specific roles, follow these guides:
- 💼 AI Engineer Path: Focuses on prompt engineering, LangChain, vector search DBs, and application APIs.
- 🛠️ ML Engineer Path: Focuses on feature engineering, model optimization, PyTorch training, and MLOps.
- 📊 Data Scientist Path: Focuses on statistics, analytical visualization, and scikit-learn models.
- 🌐 AI Researcher Path: Focuses on math, research papers, custom neural network training from scratch.
- 🚀 Freelancer & Startup Path: Focuses on Next.js, FastAPI, Stripe payment integration, building SaaS MVPs, and client pricing structures.
We have compiled comprehensive materials to help you land interviews, build resumes, and scale freelance businesses:
- 📄 Resume Writing Guide: How to format AI resumes and write XYZ bullet points
- 🎯 Interview Q&A Prep: 50+ questions and answers covering ML, DL, and System Design
- 🎓 Internship Blueprint: Timeline and cold email outreach strategy templates
- 💼 Freelancing Strategy: Upwork guidelines, finding clients, and value-based pricing
- 🌐 Open Source Guide: How to make your first Pull Request on LangChain or CrewAI
Use the checklist below to track your learning journey (python -m ai_roadmap.progress --check N):
- Milestone 1: Complete Python & Linear Algebra foundations.
- Milestone 2: Build a data wrangling script cleaning a custom CSV with Pandas.
- Milestone 3: Deployed a classification model using Scikit-Learn.
- Milestone 4: Train a custom CNN classifier using PyTorch.
- Milestone 5: Deployed a PDF Question-Answering chatbot (RAG) using Streamlit.
- Milestone 6: Build an autonomous research agent group using CrewAI.
- Milestone 7: Dockerize an API endpoint and deploy to Render cloud.
- 📝 Quizzes: 10 Q&A per phase in
docs/quizzes/(start anywhere). - 🃏 Flashcards: 40 rapid-fire cards in
docs/flashcards.md. - 📖 Docs site:
pip install -r requirements-docs.txt && mkdocs serve(strict-cleanmkdocs build).
docs/ guides + quizzes/ (70 Q&A) + flashcards + career playbooks
src/ai_roadmap/ tested library: basics → math → wrangling → features/metrics → torch → RAG → tracking
notebooks/ 01–05 worked solutions (run top-to-bottom, Colab-ready)
projects/ 01–07 runnable portfolio (each: README + requirements + CLI)
templates/ copy-paste starters: streamlit_rag, crewai_team, mcp_server, fastapi_ml, mlflow_tracking
scripts/ validate_repo, check_links, load_test
data/samples/ offline datasets (CSVs force-added, see folder README)
|
Uday Sharma Tech Creator, AI Engineer, and Growth Strategist 
|
|
If this repository helps you learn AI or advance your career, please consider giving it a Star! Your support keeps this project active and updated with the latest AI technologies.
