# ai-engineering-hub

> Open-source tutorials and project templates for LLMs, RAG, and AI agents.

ai-engineering-hub is an open-source GitHub repository containing over 90 practical AI engineering projects organized by difficulty level. It provides complete Jupyter notebooks and codebases covering OCR, RAG pipelines, model evaluations, fine-tuning, and multi-agent systems using frameworks like CrewAI and the Model Context Protocol (MCP). It is a direct reference collection for developers looking to build and deploy practical AI applications without starting from scratch.

Website: https://github.com/patchy631/ai-engineering-hub

## What it is for

- Build custom RAG pipelines over PDFs, audio, codebases, or databases
- Develop multi-agent research and automation workflows using CrewAI and MCP
- Deploy local chat and reasoning interfaces with open-source LLMs
- Fine-tune models like DeepSeek using Unsloth and Ollama

## Features

- 90+ ready-to-run projects categorized into Beginner, Intermediate, and Advanced tiers
- Implementations for basic and agentic RAG with various vector stores and embeddings
- Extensive Model Context Protocol (MCP) integrations for tools, memory, and search
- Vision and OCR setups using Llama 3.2 Vision, Gemma-3, and Qwen 2.5 VL
- Multi-agent automation workflows for content creation, research, and analysis
- Model evaluation and observability examples using CometML Opik
- Includes an end-to-end AI Engineering Roadmap for structured learning

## Pricing

Open source. Free (MIT License)

## Details

- Type: GitHub repo
- Category: [AI](https://xiry.app/topics/ai.md)
- Open source: yes
- Platforms: Linux, macOS, Windows
- For: AI engineers, data scientists, and developers building LLM and agent applications.
- Tags: llms, rag, ai agents, mcp, fine tuning, crewai, python

## Why it was saved

- [@nathanhodgson.ai, 9 February 2026](https://www.instagram.com/reel/DUi62MQDAQP/): It provides full working source code and Jupyter notebooks under an MIT license so developers can build production-ready AI applications without starting from scratch.
- [@akshay_pachaar, 23 January 2025](https://x.com/akshay_pachaar/status/1882406725313311065): A working reference for combining a web-scraping infra (Bright Data) with multi-agent orchestration (CrewAI) and a quick UI (Streamlit), and the demo shows exactly what the end product looks like.
- [@akshay_pachaar, 23 January 2025](https://x.com/akshay_pachaar/status/1882406725313311065): Shows the component breakdown and data flow of the app in one picture; a searchable pattern for building scraper + agent pipelines.

## Related

- [Inspo](https://xiry.app/resources/inspo): Design inspiration MCP server and archive for AI coding agents. (Web app, Free)
- [Agent Reach](https://xiry.app/resources/agent-reach): Open-source capability layer giving AI agents access to browse and scrape platforms (GitHub repo, Open source)
- [AutoAgent](https://xiry.app/resources/autoagent): Autonomous harness engineering framework that optimizes AI agent setups overnight (GitHub repo, Open source)
- [awesome-claude-skills](https://xiry.app/resources/awesome-claude-skills): Curated collection of skills, tools, and plugins for Claude workflows (GitHub repo, Open source)
- [Creating a company brain with Box and the Hermes Agent](https://xiry.app/resources/creating-a-company-brain-with-box-and-the-hermes-agent): Step-by-step tutorial wiring Hermes Agent to Box for cited enterprise retrieval (Article or course, Free)
- [Eliza](https://xiry.app/resources/eliza): Open-source multi-agent framework for building autonomous social and on-chain AI agents (GitHub repo, Open source)

Updated 3 October 2026. Page: https://xiry.app/resources/ai-engineering-hub

Make your own library: Xiry takes your saved reels, tweets, Reddit bookmarks and Watch Later videos, makes them searchable and pulls the useful stuff out, like this library. It's invite-only for now; join the waitlist at https://xiry.app/#waitlist.