Hi Product Hunt, I'm Loukas (Luca), a solo founder in Thessaloniki, Greece.

I built Job Opportunities API (JOA) because most job data comes from boards that copy each other, and nobody tells you when a posting has quietly closed.

What is different:

- Employer-direct: postings are read from employers' own career sites and recruiting systems (Workday, Greenhouse, SmartRecruiters and dozens of others), not from job boards. The job boards and aggregators we only read for discovery are listed by name on /provenance, and nothing from them is ever served.

- Checked at the source: postings are re-confirmed on the employer's site on a daily cycle. When a role disappears there, it leaves the live feed.

- Closed-job data: closed roles are kept with a closure date and reason, so you can study how fast roles fill, not only what is open today.

- Every field is marked published (the employer said it) or inferred (we worked it out).

- MCP for agents: Claude, Cursor and VS Code can search jobs, fetch a posting and read a company's hiring trend.

- A free key with no card: 1,000 records a month on the same data as the paid plans. With a company email, the first month of Starter or Growth is free (card needed, cancel before month two and pay nothing).

Honest limits: we are not the biggest jobs API, and coverage is strongest in North America and Europe. Pay is the weak spot: we detect a pay figure on roughly one live posting in five, and that is a lower bound.

You can check the numbers before signing up with one keyless request to /public/coverage.

What job data would you build on, and what would you need to see before you trusted a source like this? I'm here all day and I read every comment.