A complete path to becoming a self-taught AI Engineer — from first lines of code to building autonomous AI agents.
Inspired by OSSU-CS, powered by free, open-source resources.
- Why This Exists
- Curriculum Overview
- Guided Curriculum Path
- Goals
- Learning Philosophy
- How to Use This
- Project Structure
- Contributing
- License
Artificial Intelligence is reshaping every industry, yet access to quality, structured AI education is often locked behind paywalls or expensive degrees. This project offers a comprehensive, open-source curriculum for learning AI engineering, covering fundamentals to cutting-edge topics like:
- Machine Learning & Deep Learning
- Math for AI
- Natural Language Processing, Computer Vision, and RL
- Large Language Models (LLMs)
- Agentic AI, RAG, Prompt Engineering
- Real-world Projects and Deployment
Everything here is free, self-paced, and community-curated.
| Track | Topics |
|---|---|
| Programming | Python, Git, Linux, Data Structures |
| Math | Linear Algebra, Calculus, Probability, Stats |
| ML | Supervised, Unsupervised, Model Evaluation |
| Deep Learning | Neural Networks, CNNs, RNNs, Transformers |
| Specializations | NLP, Computer Vision, RL |
| Modern AI | LLMs, Agentic AI, RAG, Prompt Engineering |
| Capstone | End-to-End AI Projects |
Follow the curriculum in order, or jump to any section you’re ready for. Each link leads to detailed resources and recommendations.
| Step | Topic | Link | Est. Duration |
|---|---|---|---|
| 1 | 🛠️ Programming | Start Here | 4–6 weeks |
| 2 | 📐 Math for AI | Start Here | 6–8 weeks |
| 3 | 🤖 Machine Learning | Start Here | 6–8 weeks |
| 4 | 🧠 Deep Learning | Start Here | 4–6 weeks |
| 5 | 🔬 NLP | Natural Language Processing | 3–4 weeks |
| 👁️ Computer Vision | Computer Vision | 3–4 weeks | |
| 🧭 RL | Reinforcement Learning | 3–5 weeks | |
| 6 | 🚀 LLMs | Large Language Models | 2–3 weeks |
| ⚙️ Agentic AI | Agentic AI | 2–3 weeks | |
| 🔄 RAG | Retrieval-Augmented Generation | 1–2 weeks | |
| 🧠 Prompt Engineering | Prompt Engineering | 1–2 weeks | |
| 🧰 AI Tooling | AI Tooling | 1–2 weeks | |
| 7 | 🧪 Capstone | Capstone Projects | Ongoing |
- Build a portfolio-ready skill set for AI engineering
- Learn from world-class courses (MIT, Stanford, Fast.ai, etc.)
- Practice through projects and community feedback
- Stay current with 2025+ AI advancements
- Theory + Practice: Understand the "why", but focus on building.
- Modular: Learn in blocks; specialize when you're ready.
- Community-Driven: Suggest, improve, and contribute!
- Always Up-To-Date: We update with AI trends, not semesters.
- Start with the Foundational
- Work at your own pace — most tracks list beginner → advanced
- Track your progress in a fork or markdown file
- Build portfolio projects as you go
- Join discussions, contribute, and grow with others!
ossu-ai-engineering/
├── curriculum/ # Main curriculum files
│ ├── 01_programming.md
│ ├── 02_math.md
│ ├── ...
│ └── 07_capstone-projects.md
├── CONTRIBUTING.md # Guidelines for contributors
├── LICENSE # Open source (MIT)
└── roadmap.png # Visual overview (coming soon)
We welcome feedback, issues, and pull requests!
Check out CONTRIBUTING.md to learn how to help shape this curriculum.
- DeepLearning.AI
- Fast.ai
- Stanford CS231n: Convolutional Neural Networks for Visual Recognition
- MIT OpenCourseWare: Introduction to Deep Learning
- Papers with Code
- Hugging Face Course
- Google AI Education
- OpenAI Cookbook
This project is licensed under the MIT License.
All linked courses and materials retain their original licenses.