Job hunting is repetitive by nature — checking the same boards, re-reading postings that don't quite fit, missing the ones that do. So instead of just complaining about it, I decided to build a small tool to do it for me.
The idea: a local app that searches Swiss job platforms daily against my own profile (AI Strategy & Governance, Organizational Development, Project Portfolio Management), filters out the noise, and tells me — with a short AI-generated rationale — how well each posting actually fits.
How it's built:
Python, FastAPI, SQLite, Playwright — running entirely on my own machine, no cloud hosting, no data leaving my laptop except the job search itself and the AI scoring call. And here's the fun part: I didn't write a single line of code by hand. The whole thing was "vibecoded" — built through natural-language conversation with Claude Code, describing what I wanted and iterating from there.
What struck me most wasn't the tool itself, but how the process forced the same governance instincts I use professionally: think about data boundaries, respect platform rules, design for graceful failure. Turns out good AI governance habits apply just as well to a personal side project as to a corporate AI rollout.