Research Building a Local RAG

05 May, 2025

05-04-2025 8:37pm

Project: [[Local RAG]]
- Tasks completed:
  - Researched Local RAG tools
  - Looked at Languages
  - AI Agents
    - SmolAgents link
    - LangChain?
  - VectorDB ChromaDB

RoadMap:

Local RAG:
- There is not any local based RAG apps so think i am going to try to build a simple RAG for just analyzing personal data like Dev Log and other tasks.
Agents:
- Update: Raycast added support for ollama locally to interact with apps on mac There also is not any good local agents for working with your calendar so test building a Agent that works with apple calendars
Local Dashboard Agent Manager:
- Look at LangChain LangGraph have not dug too deep
- Check out their learning course link

Having the agent or llm to sum everything that was worked on and talked about to later run with next queries since the memory wont hold that long. This should help with accuracy for future requests.

Not sure how to build the overall app i did hear mentions of FastAPI for the backend but idk.
Connecting Ollama to these tools.
Where or how to build the app in Docker?

LangChain Academy: Learn the basics of LangGraph in our free, structured course.
Tutorials: Simple walkthroughs with guided examples on getting started with LangGraph.
Templates: Pre-built reference apps for common agentic workflows (e.g. ReAct agent, memory, retrieval etc.) that can be cloned and adapted.
How-to Guides: Quick, actionable code snippets for topics such as streaming, adding memory & persistence, and design patterns (e.g. branching, subgraphs, etc.).
API Reference: Detailed reference on core classes, methods, how to use the graph and checkpointing APIs, and higher-level prebuilt components.
Built with LangGraph: Hear how industry leaders use LangGraph to ship powerful, production-ready AI applications.