ParseHawk
github.com/parsehawk/parsehawk →Document AI that never phones home
WHAT IT SOLVES
Every time you feed a PDF to ChatGPT, there's that nagging worry — contracts, financials, internal docs, all sitting on someone else's server. ParseHawk runs 100% local by default, Ollama as the LLM backend, your data stays on your machine
WHY IT'S INTERESTING
One engine, three ways in
API for developers piping it into their stack, CLI for batch nerds, Web UI for non-technical teammates dragging and dropping. Not three products — one product serving three kinds of people
Structured JSON isn't an afterthought
Most 'document AI' tools dump a summary and call it a day. ParseHawk was built from commit #1 to spit out structured JSON — fields, types, nested relationships all programmable, ready for downstream consumption
TECH GUESS
Ollama for local LLM backend, FastAPI for the API layer, likely React/Next.js for the web UI, Python or Node for the CLI
Discussion (0)
- No comments yet — be the first.
Related
#229▶ 136Upload your textbook. AI teaches it back to you, chapter by chapter
#228▶ 112Someone needs to watch the AI spend meter
#227▶ 184Your users are kittens. Don't let them fall off the cliff