SES Group
LiveAI2026

RAG project: being able to find what I wrote myself

Retrieval across my own documents, built small enough to run on the memory the server actually has.

Role
Everything
Duration
2026, ongoing
Built with
Python · fastembed · sqlite-vec · MCP
Status
Live on my own server

The problem

A vault of notes only pays off when you can find the thing you wrote eight months ago. That requires remembering the words you used at the time, and you do not.

Approach

Memory budget first. No GPU, no large vector database, no service billed per query. Everything had to run alongside the other services on one server, and the design was chosen accordingly.

Constraint
Spare RAM on a shared server
Choice
Small embeddings, index in SQLite
Access
Exposed to tools over MCP

What it does

Indexes the vault with fastembed, stores the vectors in sqlite-vec and exposes retrieval over MCP, so an assistant can look things up in my own notes instead of guessing.

Where it stands

The retrieval is no longer a prototype. It runs as a standing service on my own server, indexes a good 1,600 documents across more than 20,000 chunks, and is used every day by AI tools that look things up over MCP instead of guessing. The code lives in its own repository with versioned releases.

Quality still varies with how well a document is written. That is not solved; it is merely in service, and the difference between the two is the reason to run it every day.

What I learned

That retrieval quality is a chunking problem long before it is a model problem. How a document is cut into pieces matters more than which model reads the pieces.

Want your own documents to be searchable?

Happy to say what worked and what cost more than it returned.