About this project

AI Engineering is an open-source e-book designed for engineers and researchers, offering a systematic guide to building production-grade AI systems ranging from Large Language Model (LLM) foundational theory to practical implementation. The project is co-maintained by the community and the Hermes Agent, with content compiled from thousands of primary sources and updated daily. The core learning path is divided into five sections: 1. **Foundations**: Understanding AI concepts, LLM principles, prompt engineering, and tool ecosystems. 2. **Engineering**: Mastering agent control loops, Harnesses, memory mechanisms, the MCP protocol, multi-agent patterns, AI-assisted programming, and RAG system design. 3. **Expertise**: Deep diving into cloud infrastructure deployment, security governance, MLOps evaluation, and data engineering. 4. **Science**: Exploring training fine-tuning, inference optimization, multimodal technologies, and embodied intelligence research. 5. **Mastery**: Discussing frontier theories, AI philosophy, safety, and future trends. Key features include: - **Traceability**: Each entry retains original source links and difficulty ratings to ensure claims are evidence-based. - **Advanced Search**: Supports local browser indexing, BM25, semantic search, and vector graph retrieval. - **AI Chat**: Integrates RAG-based chat functionality to convert site knowledge into conversational context. - **Continuous Maintenance**: Daily synchronization, quality scoring, and regression testing are performed via automated pipelines. Technically, the project uses MkDocs Material for documentation, supports local development preview via Docker, and can be deployed to Cloudflare Pages (utilizing Pages Functions, R2, and Vectorize) or GitHub Pages. The code is licensed under the MIT License, while the book content is licensed under CC BY-SA 4.0. Users can read the book online or clone the repository to build and run it locally. The project welcomes contributions such as high-quality articles, bug fixes, and optimizations for search and build processes.