AI Engineering Hub
A collection of more than 90 tutorial projects covering large language models, retrieval-augmented generation and AI agents, organized by difficulty from beginner to advanced.
The projects are grouped into beginner, intermediate and advanced tiers. Beginner projects cover single-component tasks such as local OCR apps built on vision models like Llama 3.2, Gemma-3 and Qwen 2.5 VL, local chat interfaces built on models such as DeepSeek-R1, Llama 3.2 and Gemma 3 with visible reasoning traces, basic RAG workflows using tools such as LlamaIndex, Qdrant and SambaNova, and multimodal examples such as local image generation and video chat with Gemini. Intermediate and advanced projects move on to multi-component agentic workflows and production-oriented systems, including fine-tuning.
Each project is intended to be implemented, adapted and scaled into the reader's own work, and the repository is organized with a learning roadmap for newcomers to AI engineering.
It is delivered largely as Jupyter notebooks and licensed under the MIT license.