Why your job search should be local-first

A job search produces a surprisingly intimate dataset. Put together your résumé, the roles you saved, the companies you skipped, the salary bands you filtered on, and the notes you wrote after an interview, and you have a detailed portrait of your career — including the fact that you are looking at all.

Most job-search tools collect that portrait onto their servers. JobHakken is built the other way around: local-first. The app runs on your computer, stores everything in a local SQLite database, and only talks to the outside world to do the two things it has to — fetch public job postings, and (optionally) call a language model you point it at.

What "local-first" means here

• Your data lives on your machine. Résumé, saved jobs, notes, application history, and settings are in a database file in your user directory — not in our cloud, because we do not run one for your data.
• Bring your own key (or a managed one). Résumé tailoring and keyword analysis call an LLM. You can plug in your own API key so the request goes straight from your machine to the provider, or use a managed key. Either way, your résumé is sent only when you trigger an action.
• No account required to work. The core loop — scan, score, review — runs offline against the boards you have configured.

Why it matters in practice

Say Jordan Rivera is quietly looking while still employed. With a cloud tool, "quietly" is doing a lot of work: the searches, the saved roles at competitors, the résumé versions all sit in someone else's database, subject to their retention policy, their breaches, and their business model. With a local-first tool, that history is a file Jordan controls and can delete in one move.

Privacy is the headline, but local-first buys other things too:

1. Speed. Reading and filtering hundreds of listings against a local database is instant — no round-trips.
2. Longevity. Your data is not hostage to a subscription. Cancel and the database is still yours.
3. Portability. It is a file. Back it up, move it to a new laptop, keep it forever.

The tradeoffs, honestly

Local-first is not free of tradeoffs. There is no magic cross-device sync, and the AI features need a key because the models are too large to run comfortably on a laptop for most people. We think that is the right trade: you get privacy and control, and you decide exactly when your résumé leaves the machine.

If the idea of your entire job search living in one file you own sounds good, that is the whole design.