All work

AI

DocuMind

Chat with PDFs

Cited answers in <2s

The challenge

People needed answers from large document sets without reading every page.

What I built

Upload any document set and query it in natural language. Streaming responses, semantic search and per-chunk citations with an embedded vector pipeline.

  1. 01Built an embedded vector pipeline that chunks uploads and indexes them for semantic search.
  2. 02Streamed responses so answers start appearing immediately.
  3. 03Attached per-chunk citations so every answer can be checked against the source.

The result

Cited answers in <2s

Want results like this?

Have a project, a role, or just an idea worth exploring? Drop me a message. I usually reply within a day.

Start a project xtocodex@gmail.com