Multi-ATS Jobs Scraper (Greenhouse, Lever, Ashby)

Greenhouse, Lever and Ashby each publish an open JSON feed for every company board hosted on them, and the three formats have exactly one field name in common: id. This reads a list of those boards and flattens all three into one schema, so a Stripe posting and a Ramp posting land as the same shape of row.

Those feeds are open to anyone, so there's no key to request and no anti-bot in the way. What comes back is exactly what the company's own careers page shows a visitor, read from the source instead of scraped off a rendered page.

Input

{
  "companies": ["greenhouse:stripe", "lever:spotify", "ashby:ramp"],
  "maxItems": 0
}

Each entry is ats:token. The ATS is greenhouse, lever or ashby; the token is the company's board slug, which you take straight out of their careers URL:

• Greenhouse → job-boards.greenhouse.io/<token> (the older boards.greenhouse.io/<token> and <token>.greenhouse.io forms use the same slug)
• Lever → jobs.lever.co/<company>
• Ashby → jobs.ashbyhq.com/<name>

Field	Default	Notes
companies	—	The list of boards.
ats + company	—	A convenience pair for a single board. Ignored when companies is filled.
maxItems	0	Postings per company. 0 means every open one.
proxyConfiguration	off	Optional. These APIs have no anti-bot, so a proxy only helps if you're sweeping so many boards you hit IP rate limits.
notionConnector / notionParentId	—	Optional delivery into Notion via an Apify MCP connector.

What you get per posting

{
  "ok": true,
  "ats": "greenhouse",
  "company": "stripe",
  "title": "Account Executive, AI Startups (Hunter)",
  "location": "San Francisco",
  "department": "1653 Startups - Account Executives (NA)",
  "employmentType": null,
  "url": "https://stripe.com/jobs/search?gh_jid=8130725",
  "postedAt": "2026-08-19T14:02:07-04:00",
  "description": "Who we are\n\nAbout Stripe\n\nStripe is a financial infrastructure platform…",
  "salary": null
}

description is the full posting text, HTML stripped to plain text with paragraph breaks kept. Not a snippet.

What each ATS actually publishes

The three feeds don't carry the same fields, so some values are null on a perfectly good row. Here's what each one actually exposes:

Field	Greenhouse	Lever	Ashby
employmentType	not exposed, always null	yes (its "commitment" field)	yes
salary	not exposed, always null	best effort, often null	yes, structured, still sometimes null
department	yes	department / team	department / team
postedAt	yes	yes	yes
description	yes	yes	yes

If pay data is what you're after, Ashby boards are the ones that reliably have it.

Other limits worth knowing

• Open postings only. These feeds publish what's live right now. There's no archive of closed roles.
• Rows are deduplicated by url across the whole run, so a company listed twice doesn't bill twice.
• A board that doesn't exist, or a slug typed wrong, gets its own diagnostic row and the run carries on with the rest of the list.
• department is whatever the company typed. Greenhouse boards in particular often carry internal codes like 1653 Startups - Account Executives (NA) rather than a tidy department name.

Billing

$3.00 per 1,000 jobs ($0.003 each), with no run-start fee. Flat on every plan.

One charge per posting row with ok: true. Duplicates and diagnostic rows are free, so pointing it at a list of slugs where a few turn out to be wrong only bills you for the boards that answered.