# Daily Berlin Jobs > Open-source board of tech engineering jobs in Berlin: software, data/AI, platform/DevOps/SRE, security, mobile, QA and embedded. Refreshed once a day. Free, read-only, no API key. The data is refreshed once per day, usually between 06:00 and 07:30 Europe/Berlin. Always read the status endpoint first and treat its `updated_at` as the source of truth; do not assume a fixed update time. ## For agents - [Agent instructions](https://daily-berlin-jobs.vercel.app/agents.md): a ready-made routine for a scheduled daily job check - [Status](https://daily-berlin-jobs.vercel.app/api/v1/status): last update time, job counts, next expected update window, allowed filter values - [New in the latest update (Markdown)](https://daily-berlin-jobs.vercel.app/today.md) - [All open jobs (Markdown)](https://daily-berlin-jobs.vercel.app/jobs.md) - [MCP server](https://daily-berlin-jobs.vercel.app/api/mcp): remote MCP endpoint (Streamable HTTP, no auth) with the read-only tools `get_status`, `search_jobs` and `get_job` ## API - [Jobs (JSON)](https://daily-berlin-jobs.vercel.app/api/v1/jobs): query parameters `since` (ISO 8601; jobs first seen after this time), `latest=1` (jobs added in the most recent update), `role`, `level`, `work_mode` (comma-separated, values listed in status `filters`), `q` (free text over title, company and tech stack), `source` (`ats` or `linkedin`), `limit` (1-500, default 100), `cursor` (from `next_cursor`), `format=md` - Job detail: `https://daily-berlin-jobs.vercel.app/api/v1/jobs/{id}` (JSON) or `https://daily-berlin-jobs.vercel.app/jobs/{id}.md` (Markdown) - [RSS feed](https://daily-berlin-jobs.vercel.app/feed.xml) - [Companies](https://daily-berlin-jobs.vercel.app/api/v1/companies): every company crawled directly from its own career site, with website, ATS, career page, source health and open job count (Markdown: https://daily-berlin-jobs.vercel.app/companies.md) ## Notes - Jobs are removed 30 days after posting. - Full descriptions are included only for jobs that come from a company's own career site (`source: "ats"`). Jobs with `source: "linkedin"` include metadata and a link only. - Responses are cached and change about once a day; a few requests per run are enough. - Source code and issues: https://github.com/umutyesildal/daily-berlin-jobs