AI Open Knowledge is a community-driven, open-source project dedicated to making Artificial Intelligence knowledge accessible to everyone. We bring together learning resources, engineering guides, prompts, workflows, AI agents, tutorials, research papers, datasets, examples, templates, best practices, and much more—all in one place.
Whether you're taking your first steps in AI or advancing the state of the art, you'll find resources to learn, build, and share. Every contribution helps expand a growing knowledge base that benefits learners, practitioners, and researchers around the world.
🌍 Free for everyone. Open to everyone. Built by everyone.
Our community includes students, developers, AI engineers, researchers, educators, organizations, and open-source enthusiasts who believe that knowledge grows when it's shared.
⭐ If you find this repository useful, please consider giving it a star and contributing. Every pull request, resource, correction, or idea helps make AI knowledge more accessible for everyone. Together, we're building the future of open AI knowledge.
- 🧠 AI Open Knowledge
- ❓ Why AI Open Knowledge?
- ✨ Features
- 🗂️ Repository Structure
- 🧭 Complete Repository Navigation
- 📚 Documentation
- 🎓 Learning Areas
- 🛤️ Learning Paths
- 👥 Who Is This Repository For?
- 🚀 How To Use
- 🤝 Contribution
- 🌐 Community Guidelines
- 🎯 Repository Goals
- 🗺️ Roadmap
- 💬 FAQ
- 🏆 Best Practices
- 🧭 Repository Philosophy
- 📈 Recommended Order of Learning
- 🔮 Future Vision
- 🆘 Support
- 🔒 Security
- 🏛️ Governance
- 📝 Changelog
- ⚖️ License
- 🙏 Acknowledgements
- ⭐ Star History
- 👨💻 Contributors
Artificial Intelligence is evolving faster than any technology in history — but the knowledge needed to keep up is scattered, fragmented, and often locked behind paywalls.
| Problem | Reality |
|---|---|
| 🧩 Fragmented resources | Great content exists — but it's spread across thousands of blogs, papers, videos, courses, and repos with no unified structure. |
| ⚡ Rapid AI evolution | Techniques become outdated within months. LLMs, agents, RAG, and MCP move faster than any single curriculum can track. |
| 🌫️ Knowledge scattered | Beginners don't know where to start; practitioners waste hours re-discovering what others already documented. |
| 💰 Paywalls and gatekeeping | High-quality AI education is often expensive, excluding students and developers in many parts of the world. |
| 🏝️ Isolated learning | Most people learn AI alone, without a community to review, discuss, and improve their understanding. |
AI Open Knowledge solves this through:
- 🤝 Community collaboration — hundreds of contributors keep content accurate, current, and battle-tested.
- 🔓 Open source — everything is transparent, versioned, reviewable, and forever free.
- 🗂️ Organized structure — a single, well-designed taxonomy covering the entire AI landscape.
- 🔄 Continuous updates — pull requests keep pace with the field, so knowledge never goes stale.
- 🌍 Global accessibility — anyone with an internet connection can learn, use, and contribute.
Tip
Think of this repository as a living encyclopedia of practical AI knowledge — maintained like software, versioned like code, and open like the web should be.
|
|
|
| Folder | Description | Link |
|---|---|---|
| 🤖 Agents | Agent architectures, orchestration patterns, multi-agent systems | ./agents/ |
| ⚙️ AI Engineering | Designing, building, and shipping production AI systems | ./ai-engineering/ |
| 🧱 AI Fundamentals | Learn AI basics — concepts, history, and core ideas | ./ai-fundamentals/ |
| 🛡️ AI Safety | Alignment, ethics, responsible AI, and risk mitigation | ./ai-safety/ |
| 🔐 AI Security | Prompt injection, jailbreaks, red-teaming, and defenses | ./ai-security/ |
| 🧰 AI Tools | Tooling landscape, comparisons, and usage guides | ./ai-tools/ |
| 🌟 Awesome Lists | Curated awesome-style collections of external resources | ./awesome-lists/ |
| 📊 Benchmarks | Model benchmarks, leaderboards, and comparison data | ./benchmarks/ |
| 📚 Books | Recommended books and structured reading lists | ./books/ |
| 💼 Career | AI career paths, roles, skills, and growth strategies | ./career/ |
| 📝 Cheatsheets | Quick-reference sheets for tools, math, and frameworks | ./cheatsheets/ |
| 🌐 Community | Community spaces, events, and collaboration channels | ./community/ |
| 💻 Computer Science | CS foundations: algorithms, data structures, systems | ./computer-science/ |
| 👁️ Computer Vision | Image classification, detection, segmentation, and more | ./computer-vision/ |
| 🔧 Data Engineering | Data pipelines, ETL, quality, and preparation for AI | ./data-engineering/ |
| 📂 Datasets | Curated dataset catalogs across domains | ./datasets/ |
| 🧠 Deep Learning | Neural networks, architectures, and training techniques | ./deep-learning/ |
| 📏 Evaluation | Evals, metrics, and testing methodologies for AI systems | ./evaluation/ |
| 💡 Examples | Runnable, practical code examples | ./examples/ |
| ❔ FAQ | Frequently asked questions across all topics | ./faq/ |
| 🎛️ Fine-Tuning | SFT, LoRA, RLHF, and model adaptation recipes | ./fine-tuning/ |
| 🏗️ Frameworks | ML/AI frameworks, libraries, and ecosystem guides | ./frameworks/ |
| 🔮 Future Topics | Emerging and frontier AI research areas | ./future-topics/ |
| 🎨 Generative AI | Image, video, audio, and multimodal generation | ./generative-ai/ |
| 📕 Glossary | AI terminology explained in plain language | ./glossary/ |
| 🖥️ Infrastructure | GPUs, clusters, MLOps, and compute infrastructure | ./infrastructure/ |
| 🎤 Interview | Interview questions, answers, and preparation guides | ./interview/ |
| 🛤️ Learning Paths | Structured curricula by level and role | ./learning-paths/ |
| 🗣️ LLMs | Large language models: architecture, training, usage | ./llms/ |
| 📈 Machine Learning | Classical ML algorithms, theory, and practice | ./machine-learning/ |
| ➗ Mathematics | Linear algebra, calculus, probability, statistics | ./mathematics/ |
| 🔌 MCP | Model Context Protocol servers, clients, and patterns | ./mcp/ |
| 🚀 Model Serving | Inference optimization, deployment, and scaling | ./model-serving/ |
| 💬 NLP | Natural language processing tasks and techniques | ./nlp/ |
| 📄 Papers | Key research papers with summaries and notes | ./papers/ |
| 👨💻 Programming | Programming skills and languages for AI work | ./programming/ |
| 🏗️ Projects | End-to-end project ideas, builds, and walkthroughs | ./projects/ |
| ✍️ Prompt Engineering | Prompting techniques, patterns, and science | ./prompt-engineering/ |
| 💬 Prompts | Ready-to-use, categorized prompt library | ./prompts/ |
| 📡 RAG | Retrieval-augmented generation pipelines and patterns | ./rag/ |
| 🕹️ Reinforcement Learning | RL theory, algorithms, and applications | ./reinforcement-learning/ |
| 🔬 Research | Research methods, directions, and open problems | ./research/ |
| 🔗 Resources | External resources, links, and collections | ./resources/ |
| 🦾 Robotics | Robotics, embodied AI, and control systems | ./robotics/ |
| 🧩 Skills | Reusable AI skill definitions and libraries | ./skills/ |
| 🔊 Speech | ASR, TTS, and speech processing | ./speech/ |
| 📋 Templates | Project, document, and workflow templates | ./templates/ |
| 📖 Tutorials | Step-by-step guides across all levels | ./tutorials/ |
| 🗄️ Vector Databases | Embeddings, vector stores, and similarity search | ./vector-databases/ |
| 🔄 Workflows | AI workflow patterns, automation, and orchestration | ./workflows/ |
| Document | Purpose | Link |
|---|---|---|
| 📖 README | Project overview and navigation (this file) | README.md |
| 🤝 Contributing Guide | How to contribute content, fixes, and ideas | CONTRIBUTING.md |
| 🌐 Code of Conduct | Community behavior standards | CODE_OF_CONDUCT.md |
| 🔒 Security Policy | Reporting vulnerabilities responsibly | SECURITY.md |
| 🆘 Support Guide | Where and how to get help | SUPPORT.md |
| 🗺️ Roadmap | Planned direction and milestones | ROADMAP.md |
| 🏛️ Governance | Decision-making and maintainer structure | GOVERNANCE.md |
| 📝 Changelog | Notable changes over time | CHANGELOG.md |
| ⚖️ License | Usage rights and terms | LICENSE |
🧠 Core AI & ML
| Area | Start Here |
|---|---|
| 🧱 Artificial Intelligence | ./ai-fundamentals/ |
| 📈 Machine Learning | ./machine-learning/ |
| 🧠 Deep Learning | ./deep-learning/ |
| 🕹️ Reinforcement Learning | ./reinforcement-learning/ |
🗣️ LLMs & Generative AI
| Area | Start Here |
|---|---|
| 🗣️ LLMs | ./llms/ |
| ✍️ Prompt Engineering | ./prompt-engineering/ |
| 🤖 AI Agents | ./agents/ |
| 📡 RAG | ./rag/ |
| 🔌 MCP | ./mcp/ |
| 🎨 Generative AI | ./generative-ai/ |
| 🎛️ Fine-Tuning | ./fine-tuning/ |
👁️ Perception & Interaction
| Area | Start Here |
|---|---|
| 👁️ Computer Vision | ./computer-vision/ |
| 💬 NLP | ./nlp/ |
| 🔊 Speech | ./speech/ |
| 🦾 Robotics | ./robotics/ |
🔬 Research & Foundations
| Area | Start Here |
|---|---|
| 🔬 Research | ./research/ |
| 📄 Papers | ./papers/ |
| 👨💻 Programming | ./programming/ |
| ➗ Mathematics | ./mathematics/ |
| 💻 Computer Science | ./computer-science/ |
🏗️ Engineering & Operations
| Area | Start Here |
|---|---|
| ⚙️ AI Engineering | ./ai-engineering/ |
| 🖥️ Infrastructure | ./infrastructure/ |
| 🚀 Model Serving | ./model-serving/ |
| 🔧 Data Engineering | ./data-engineering/ |
| 🗄️ Vector Databases | ./vector-databases/ |
| 📏 Evaluation | ./evaluation/ |
🛡️ Trust, Safety & Growth
| Area | Start Here |
|---|---|
| 🔐 Security | ./ai-security/ |
| 🛡️ Safety | ./ai-safety/ |
| 💼 Career | ./career/ |
| 🔗 Resources | ./resources/ |
All curated paths live in ./learning-paths/. Choose by level or by role:
| Level | Focus | Suggested Route |
|---|---|---|
| 🌱 Beginner | What AI is, core concepts, first models | Fundamentals → Math → Programming → ML |
| 🌿 Intermediate | Deep learning, NLP, real projects | Deep Learning → NLP → Projects → Examples |
| 🌳 Advanced | LLMs, agents, RAG, production systems | LLMs → RAG → Agents → AI Engineering |
| 🏔️ Expert | Research, fine-tuning, infrastructure at scale | Fine-Tuning → Papers → Infrastructure → Research |
| Role | Core Track |
|---|---|
| ⚙️ AI Engineer | AI Engineering → LLMs → RAG → Evaluation → Model Serving |
| 📈 ML Engineer | Machine Learning → Deep Learning → Data Engineering → Infrastructure |
| 🗣️ LLM Engineer | LLMs → Prompt Engineering → Fine-Tuning → Vector Databases → Evaluation |
| 🤖 Agent Engineer | Agents → MCP → Workflows → Skills → AI Security |
| 🔬 Research Scientist | Mathematics → Papers → Research → Benchmarks → Future Topics |
| Audience | What You'll Find |
|---|---|
| 🎓 Students | Free structured curricula, math foundations, glossaries, and beginner-friendly paths — no tuition required. |
| 👨💻 Developers | Practical examples, templates, cheat sheets, and engineering patterns you can apply today. |
| 🔬 Researchers | Paper collections, benchmarks, open problems, and research methodology guides. |
| 🧑🏫 Teachers & Educators | Ready-to-use teaching materials, learning paths, and curriculum structures. |
| 🏢 Companies | Best practices, security guidelines, evaluation frameworks, and production architectures. |
| 🚀 Startups | Fast-track knowledge for building AI products: RAG, agents, serving, and tooling. |
| 🤝 Open-Source Contributors | A welcoming project where documentation contributions genuinely matter. |
Tip
You don't need to read everything. Pick your entry point based on who you are and what you need.
-
🌱 New to AI? Start at
./ai-fundamentals/, then follow the Beginner path in./learning-paths/. Keep the./glossary/open in another tab. -
🎯 Have a specific goal? Use the Complete Repository Navigation table to jump directly to your topic — for example
./rag/if you're building a retrieval pipeline. -
🛠️ Building something right now? Grab a starting point from
./templates/or./examples/, and check./workflows/and./prompts/for reusable pieces. -
📚 Studying for interviews? Head to
./interview/and reinforce weak areas with./cheatsheets/. -
🔍 Looking for something specific? Use GitHub's built-in search (/ on the repo page) — every document is plain Markdown and fully searchable.
-
⭐ Want to stay updated? Star and watch the repository, and check
CHANGELOG.mdfor notable additions.
Contributions are the heart of this project. Whether you fix a typo, add a tutorial, translate content, or design a whole new learning path — you make AI knowledge more accessible for everyone.
📘 Read the full guide: CONTRIBUTING.md
# 1. Fork the repository on GitHub
# 2. Clone your fork
git clone https://github.com/YOUR-USERNAME/ai-open-knowledge.git
cd ai-open-knowledge
# 3. Create a feature branch
git checkout -b add/rag-evaluation-guide
# 4. Make your changes (add or improve content)
# 5. Commit with a clear message
git commit -m "docs(rag): add evaluation guide for RAG pipelines"
# 6. Push and open a Pull Request
git push origin add/rag-evaluation-guide- ✍️ New guides, tutorials, and explanations
- 🔗 High-quality curated resources
- 💬 Prompts, skills, and agent patterns
- 🐛 Fixes for outdated or incorrect content
- 🌍 Translations
- 💡 Ideas — open an issue to discuss before writing
Important
All contributions are reviewed for accuracy, clarity, and neutrality before merging. See CONTRIBUTING.md for content standards.
We are committed to a welcoming, respectful, and inclusive community. All participants — contributors, maintainers, and users — are expected to follow our Code of Conduct.
In short:
- 🤝 Be respectful and constructive
- 🌍 Welcome newcomers and different perspectives
- 🎯 Focus on the content, not the person
- 🚫 No harassment, discrimination, or gatekeeping
| Horizon | Goals |
|---|---|
| 🟢 Short Term | Populate every folder with high-quality core content; establish contribution standards; build the initial maintainer team. |
| 🟡 Mid Term | Complete learning paths for all roles and levels; grow an active contributor community; add multilingual content; automate quality checks with CI. |
| 🔵 Long Term | Become the default free reference for AI education and engineering; cover every major AI subfield with depth; partner with educators and open-source communities worldwide. |
| 🌟 Ultimate Vision | The world's largest free and open AI knowledge repository — a universal, community-maintained public good that keeps pace with AI itself. |
The detailed roadmap — including planned sections, milestones, and priorities — lives in ROADMAP.md.
Want to influence the direction? Open an issue with your proposal or comment on existing roadmap discussions.
1. Is this really free?
Yes — completely and permanently. All content is open source under the terms of the LICENSE. No paywalls, no premium tiers, no sign-ups.
2. I'm a complete beginner. Where do I start?
Start with ./ai-fundamentals/ and follow the Beginner path in ./learning-paths/. Keep the ./glossary/ handy for unfamiliar terms.
3. How is content kept up to date?
Through the community: contributors submit pull requests when techniques evolve, and maintainers periodically review high-traffic sections. Outdated content can be flagged via issues.
4. Can I use this content in my courses, blog, or company training?
Yes, in accordance with the LICENSE. Attribution is appreciated. Educators are explicitly encouraged to build on this material.
5. How do I contribute if I've never contributed to open source before?
Perfect — this is a great first project! Read CONTRIBUTING.md, look for issues labeled good first issue, and start with something small like fixing a typo or adding a resource link.
6. Do contributions need to be in English?
English is the primary language, but translations are welcome and encouraged. See CONTRIBUTING.md for translation guidelines.
7. Is this repository vendor-neutral?
Yes. We cover tools and models from all providers — open-source and commercial — objectively, based on evidence and community experience. No paid placements, ever.
8. What's the difference between prompts/ and prompt-engineering/?
./prompt-engineering/ teaches the techniques and science of prompting. ./prompts/ is a library of ready-to-use prompts you can copy directly.
9. What is MCP and why does it have its own folder?
The Model Context Protocol is an open standard for connecting AI models to tools and data sources. It's becoming foundational for agent systems, so ./mcp/ covers servers, clients, and integration patterns in depth.
10. Can I request a topic that doesn't exist yet?
Absolutely — open an issue describing the topic and why it matters. If it fits the taxonomy, it will be added to the ROADMAP.md or you can contribute it yourself.
11. How is content quality ensured?
Every pull request goes through review for accuracy, clarity, sourcing, and neutrality. CI checks validate links and formatting. Community feedback via issues catches anything that slips through.
12. Does this repository include code, or only documentation?
Both. Most content is Markdown documentation, but ./examples/, ./projects/, and ./templates/ contain runnable code and starter kits.
13. How do I report incorrect or outdated information?
Open an issue with the file path and a description of the problem — or better, submit a pull request with the fix.
14. How do I report a security concern?
Please follow the responsible disclosure process described in SECURITY.md. Do not open public issues for security vulnerabilities.
15. Who maintains this project and how are decisions made?
The project is governed by a team of maintainers with community input. Roles, responsibilities, and decision-making processes are documented in GOVERNANCE.md.
Content in this repository follows — and teaches — these principles:
- ✅ Cite sources — claims link to papers, docs, or reproducible evidence
- ✅ Show, don't just tell — concepts come with examples and code where possible
- ✅ Stay current — dated content is marked and refreshed
- ✅ Start simple — every topic offers a beginner-accessible entry point
- ✅ Be practical — theory is always connected to real-world application
- ✅ Remain neutral — tools and models are compared on evidence, not hype
- ✅ Think security-first — engineering content addresses safety and security implications
- ✅ Keep it maintainable — consistent structure and naming across all folders
Find dedicated best-practice guides throughout the repository, especially in ./ai-engineering/, ./ai-security/, and ./evaluation/.
| Principle | What It Means |
|---|---|
| 🔓 Open | All content is public, versioned, and transparently maintained. Anyone can inspect how knowledge is created and reviewed. |
| 💸 Free | Knowledge here will never be behind a paywall. Access to AI education is a right, not a privilege. |
| 🤝 Collaborative | The community writes, reviews, and improves everything together. No single voice owns the truth. |
| 🔬 Evidence-Based | Claims are backed by papers, benchmarks, and reproducible results — not marketing or hype. |
| ⚖️ Vendor Neutral | We cover the entire ecosystem fairly. No paid placements, no favoritism, no hidden agendas. |
A pragmatic route from zero to advanced AI practitioner:
- 🧱 Foundations —
./ai-fundamentals/+./glossary/— understand what AI is and speak the language - ➗ Mathematics —
./mathematics/— linear algebra, probability, and calculus essentials - 👨💻 Programming —
./programming/+./computer-science/— coding skills for AI work - 📈 Machine Learning —
./machine-learning/— classical algorithms and core theory - 🧠 Deep Learning —
./deep-learning/— neural networks and modern architectures - 💬 NLP —
./nlp/— language understanding foundations - 🗣️ LLMs —
./llms/— how large language models work and how to use them - ✍️ Prompt Engineering —
./prompt-engineering/+./prompts/— communicate with models effectively - 📡 RAG —
./rag/+./vector-databases/— ground models in your own data - 🤖 Agents —
./agents/+./mcp/+./workflows/— build autonomous AI systems - 🎛️ Fine-Tuning —
./fine-tuning/— adapt models to your domain - ⚙️ AI Engineering —
./ai-engineering/+./evaluation/+./model-serving/— ship to production - 🛡️ Safety & Security —
./ai-safety/+./ai-security/— build responsibly - 🔬 Research & Beyond —
./papers/+./research/+./future-topics/— push the frontier
Note
This order is a suggestion, not a rule. Skip ahead if you already have the prerequisites — every section stands on its own.
We believe the future of AI education should be open, free, and community-owned.
Our vision is for AI Open Knowledge to become one of the largest AI knowledge bases in the world — the place millions of people go to learn AI, the reference engineers consult when building systems, and the foundation educators build their courses upon.
As AI reshapes every industry, the gap between those who understand it and those who don't will define opportunity for a generation. This repository exists to close that gap — for every student without tuition money, every developer in an underserved region, every researcher without institutional access, and every curious mind anywhere on Earth.
The destination: a living, universal, always-current encyclopedia of AI — maintained like great software, trusted like great science, and open like the best of the internet.
Join us in building it. 🚀
Need help? Check SUPPORT.md for:
- 📖 Where to find answers (
./faq/is a great start) - 💬 How to ask questions via issues and discussions
- 🌐 Community channels (
./community/)
Found a vulnerability or a security concern in any content or example code?
Please follow the responsible disclosure process in SECURITY.md. Do not report security issues in public issues.
Project structure, maintainer roles, decision-making processes, and how to become a maintainer are documented in GOVERNANCE.md.
All notable changes to the repository are tracked in CHANGELOG.md, following Keep a Changelog conventions.
This project is licensed under the terms described in LICENSE.
You are free to use, share, and build upon this knowledge in accordance with the license terms.
This project stands on the shoulders of the global open-source and AI research communities.
- 💖 To every contributor — your pull requests, issues, reviews, and ideas are what make this repository alive. Thank you.
- 🌍 To the open-source community — for proving that the best knowledge is built together, in the open.
- 🔬 To researchers and educators — for publishing openly and teaching generously.
- 🏗️ To the maintainers of the tools, frameworks, and standards we document — your work makes modern AI possible.
Thanks go to these wonderful people:
Contributions of any kind are welcome — see CONTRIBUTING.md to add yourself to this list!
Every star raises visibility. Every contribution raises quality. Every reader raises the impact.
⭐ Star this repository • 🍴 Fork it • 🤝 Contribute • 💬 Join the community
Made with ❤️ by the AI Open Knowledge community
Free forever. Open to everyone. Built together.