Skip to content

Sample Projects

Hands-on projects that wire together every concept covered in this portal into a working, end-to-end RAG application you can run on your own machine.

What you'll learn

  • How individual building blocks (loaders, embeddings, vector stores, LLMs) combine into a real system
  • Common patterns for structuring a RAG codebase
  • How to extend and adapt a working baseline to your own data and use-case

Available projects

Project Stack Difficulty What you learn
Local RAG sentence-transformers · ChromaDB · Ollama · Streamlit Beginner End-to-end local pipeline, chunking strategy, Streamlit chat UI
Multi-PDF Research Assistant pypdf · ChromaDB · Ollama · Streamlit Intermediate Multi-document ingestion, page-aware metadata, per-file/page citations
Codebase Q&A sentence-transformers · ChromaDB · Ollama Intermediate Walking a repo, line-based chunking, path:line citations
Hybrid + Rerank App BM25 · ChromaDB · cross-encoder · Ollama Advanced Hybrid search, RRF fusion, cross-encoder reranking, stage-by-stage comparison

How to use these projects

Each project page includes:

  1. An architecture diagram
  2. A file-by-file code walkthrough
  3. Step-by-step setup and run instructions
  4. Config knobs you can tune
  5. Ideas for extending the project further

All projects run entirely on your local machine — no API keys or cloud accounts required.

New to RAG?

Work through the tutorials before tackling a full project. The tutorials introduce each concept in isolation so the project code makes sense at a glance.

Next steps