Job Match Finder

Finding suitable IT roles across Switzerland meant repeatedly searching job boards, recruiter listings and individual employer websites. The same vacancy often appeared several times; many results were irrelevant to a systems administration, Microsoft 365 and infrastructure background; and agencies frequently hid the real employer.

Job boards can also remove adverts without making it clear whether the vacancy actually closed. Manually checking hundreds or thousands of employers is unrealistic — and many good employers advertise only on their own careers pages, so they never appear on aggregators at all.

Off-the-shelf job alerts still centre on job-board searches. I needed something different: monitor employers directly, keep them on a watchlist even when they have no open roles today, and score vacancies against my own skills and preferences.

So I used AI-assisted development to build a self-hosted Job Match Finder that starts from employer discovery rather than from a single job board.

The long-term design is not “search a job board again”. It is company registry → employer → official website → careers page → recruiting system → continuous monitoring → vacancy → profile matching. Employers stay monitored so new roles can be found later.

I defined the matching logic, the employer-discovery workflow, the validation rules and the safety requirements: verify company identity before trusting a website, validate new sources before treating them as authoritative, and keep provenance for each vacancy.

Cursor was used to implement, test and iteratively improve the application against real vacancy and employer data. AI helped build the tool — the match scores and ranking decisions follow rules I set.