Greenhouse Jobs Scraper - Every Opening From Any Greenhouse Board

A company that is hiring is a company that is spending. Pull every live opening from any Greenhouse board via the official public API, with departments joined in. Paste a company domain - no board token to look up.

---

What it does

Give it a list of Greenhouse board tokens. It calls the official public Greenhouse board API for each one and returns every live opening in a flat, spreadsheet-ready schema.

This reads the API Greenhouse already publishes for its customers' own careers pages. That means no proxy, no headless browser, no anti-bot arms race — and nothing that silently breaks when a page layout changes.

Greenhouse omits departments from its jobs endpoint. This Actor also reads the departments endpoint and joins them in, so department is populated where most Greenhouse scrapers leave it empty.

---

Input

{
  "boards": ["stripe.com", "https://job-boards.greenhouse.io/anthropic"],
  "titleKeywords": ["engineer"],
  "locations": ["Remote"],
  "postedWithinDays": 7
}

The board token is the company slug in job-boards.greenhouse.io/ — but you rarely need to look it up. Paste the company domain (acme.com, www.acme.com) or the full careers URL and the Actor pulls the company name out for you.

Field	Default	Notes
boards	—	Required. Tokens or full careers URLs.
titleKeywords	—	Keep only titles containing any of these.
excludeKeywords	—	Drop titles or departments matching any of these.
locations	—	Keep only matching locations.
departments	—	Matches department or team.
remoteOnly	false	Remote positions only.
postedWithinDays	0	0 = no limit. Set 1 for a daily new-jobs feed.
includeDescription	false	Adds full description text. Much larger results.
onlyNewSinceLastRun	false	Return only what opened or closed since the last run.
maxJobsPerBoard	0	0 = no limit. The console starts at 10 so a first trial run stays cheap.
concurrency	5	Boards fetched in parallel.

---

Output

One row per opening:

Field	Description
companyName, boardToken, ats	Who, and where it was read from
jobId	Stable ID on Greenhouse
title	Job title
department, team	Org placement
employmentType	Full-time, contract, intern…
location, isRemote, workplaceType	Where the work happens
salary	Compensation range, when published
publishedAt, updatedAt	ISO 8601 timestamps
jobUrl, applyUrl	Public posting and application links
globalId	{ats}:{token}:{jobId} — stable key for joining and deduping
description	Full text, only when requested

Export as Excel, CSV, JSON or XML, or pull it through the API.

A board that cannot be found returns one row with an error and a hint, so a bad token never silently vanishes from your results.

---

Monitoring: only what changed

Set onlyNewSinceLastRun: true and schedule it. The first run records a baseline and returns everything. Every run after that returns only:

• jobs that opened since your last run — isNew: true
• jobs that have since closed — isClosed: true

A morning with no hiring activity returns nothing at all and costs only the start fee. You are never billed for re-reading a job you already saw.

Two things worth knowing:

• Changing the boards or filters starts a fresh baseline. Otherwise every job you stopped asking about would be reported as newly closed, which is a wrong answer rather than a noisy one.
• A closed row carries what was recorded when the job was last seen — company, title and link. Once a posting is gone there is nothing left to re-read, so the remaining fields are omitted rather than shown stale.

Without monitoring

postedWithinDays: 1 also gives you a daily feed, based on each board's own posted date. It is simpler, but it cannot tell you when a role closed, and it depends on the company having set a date at all. Use onlyNewSinceLastRun if you care about either.

---

Quality notes

• Timestamps are normalised to ISO 8601, so date filters behave predictably.
• Postings that cannot be read are skipped and reported in the log rather than quietly dropped — and you are not billed for them.
• One malformed posting never takes down the rest of the board.

---

Limits

• Greenhouse boards only. For a company on a different platform, see the multi-platform version of this Actor.
• Boards set to private or password-protected are not accessible.
• department and team are only as good as what the company fills in.

---

The endpoint this reads

https://boards-api.greenhouse.io/v1/boards/<company>/jobs

That is Greenhouse's own public board API - the one that fills its customers' careers pages. It needs no key, no login and no proxy, which is why this Actor does not use any.

---

FAQ

Does Greenhouse have a public jobs API? Yes, for job boards. The endpoint above returns the live postings for one company as JSON. It is public because the careers page itself is public.

Do I need an API key, a login, or a proxy? No. None of the three.

Where do I find the board token? You do not have to. Paste the company's domain, the URL of any job posting, or the identifier itself - stripe.com is a working example - and this Actor resolves it. If you already know the token - the company slug in job-boards.greenhouse.io/

Why would I use this instead of calling the endpoint myself? For one company, you probably should not - it is one HTTP request, and this README just told you the URL. This Actor earns its keep at the point where that stops being true: dozens of companies at once, tokens you would otherwise have to look up by hand, a flat schema shared across six ATS platforms, a globalId you can join on, and a mode that returns only what changed since the last run instead of the whole board every morning.

Can I get only the jobs that opened or closed since last time? Yes. Set onlyNewSinceLastRun to true and each run returns the new postings, plus a row for every posting that disappeared. See the monitoring section above.

What does a run cost? You pay per row returned - about a tenth of a cent each - plus a fraction of a cent to start the run. A company with 40 openings lands around five cents. With onlyNewSinceLastRun on, a scheduled daily run usually returns a handful of rows rather than the whole board.

A company is not found. Why? Either it is not on Greenhouse - most large enterprises use Workday, Taleo, iCIMS or SuccessFactors - or its board is private. The run tells you which, per company, in a row you can read rather than a silent gap.

---

Support

A board that will not resolve? Open an issue on the Issues tab with the careers URL.