BAWS Research Companion

A tool built out of real workflow friction — keeping reading, notes and experiments in one place while doing research.

YEAR
2025
STACK
python
STATUS
WIP
SOURCE
REPOSITORY ↗

overview

BAWS Research Companion is a working tool for managing a research workflow: collecting sources, keeping notes connected to what produced them, and tracking the small experiments that research actually consists of.

motivation

This one started from annoyance, not curiosity. Research material fragments across PDFs, browser tabs, notebooks and half-remembered terminal sessions — and the connections are the research. I wanted a tool shaped around how I actually work rather than how a product assumes I do, which meant building it.

the problem being explored

  • What the minimum structure of research state is — enough to be useful, not so much that maintaining it becomes a second job.
  • How to keep provenance intact: which run, which source, which version.
  • Where automation helps a workflow and where it silently adds friction.

approach

Plain files over databases, so everything stays greppable, portable and version-controlled. Small commands over a monolithic app, so the tool fits into a terminal workflow instead of replacing it. It is deliberately unfinished in places — it grows when the workflow does.

what i learned

  • Tool-building is systems design under the constraint "one user, me" — every abstraction must pay rent immediately.
  • The best data model is the one you'll actually maintain at 1 a.m.
  • Building for yourself first surfaces requirements no specification would have mentioned.

repository

Source lives on GitHub.