ATS Job Scraper — Greenhouse, Lever, Ashby, Workday & more

Scrape job postings straight from company career boards instead of from a job aggregator. Paste the board URL of any company you want watched and this Actor detects which applicant tracking system they run, fetches the postings, and returns them all in one shape.

Eight systems are supported:

Applicant tracking system	Typical board URL
Greenhouse	https://boards.greenhouse.io/stripe
Lever	https://jobs.lever.co/netflix
Ashby	https://jobs.ashbyhq.com/ramp
Workable	https://apply.workable.com/persado
SmartRecruiters	https://jobs.smartrecruiters.com/Visa
Rippling	https://ats.rippling.com/rippling/jobs
Breezy HR	https://acme.breezy.hr
Workday	https://nvidia.wd5.myworkdayjobs.com/en-US/NVIDIAExternalCareerSite

You can also write greenhouse:stripe or lever:netflix if you already know the platform.

Why scrape career boards rather than a job site

Postings appear on a company's own board first and reach the aggregators later, with fields dropped along the way. Going to the source gives you the department, the team, the employment type, and the real application link, on the day it goes live.

What makes this one different

Only-new delivery. The Actor remembers what it already returned. Schedule it daily and each run gives you what appeared since yesterday, not the same 600 rows every morning. A run that finds nothing new costs nothing beyond platform compute.

One dead board does not kill the run. Companies rename and close boards constantly. Every board is fetched independently; a 404 is logged and skipped and you still get results from the rest. Watching 200 companies is normal use here.

Correct billing on multi-location roles. Rippling publishes one row per location — its own board returns 691 rows for 367 actual jobs. Those are collapsed into one posting with a list of locations, so you are not charged for the same role four times. This is checked by a test.

Workday is included. Large employers run Workday, and it needs a different request shape from every other board. Most career-board Actors skip it.

Output

One record per posting:

{
  "uid": "143fa3766650b2e1",
  "platform": "ashby",
  "company": "ramp",
  "company_name": null,
  "source_id": "34413f8d-26bf-4bbc-8ade-eb309a0e2245",
  "title": "Security Engineer, Cloud",
  "location": "New York, NY (HQ)",
  "locations": ["New York, NY (HQ)", "Remote (US)"],
  "remote": true,
  "department": "Engineering",
  "team": "Backend",
  "employment_type": "FullTime",
  "salary": "$180K – $220K",
  "posted": "2026-05-01",
  "updated": null,
  "description": "Secure the cloud …",
  "url": "https://jobs.ashbyhq.com/ramp/34413f8d-26bf-4bbc-8ade-eb309a0e2245"
}

Filters

• keywords / excludedKeywords — match against title, department, and team
• locations — plain text, so London and United Kingdom both work
• departments — match against department and team
• remoteOnly — uses the board's own remote flag where there is one, the location text otherwise
• postedWithinDays — publication window
• onlyNew — only what has appeared since the last run (on by default)
• includeDescription — drop the body if you only need titles and links
• maxResults — hard cap on postings delivered, and therefore on cost

What each board actually publishes

Boards differ, and pretending otherwise would make the output look complete when it is not:

Platform	Description in output	Publication date	Salary
Greenhouse	yes	yes	no
Lever	yes	yes	no
Ashby	yes	yes	when the employer opts in
Breezy	when set	yes	when set
Workable	no	yes	no
SmartRecruiters	no	yes	no
Rippling	no	no	no
Workday	no	no	no

Rippling and Workday publish no dates at all, so postedWithinDays cannot filter them and their postings are always kept. Use onlyNew to get recency on those boards — that is exactly what it is for.

Use cases

• Job boards and aggregators — pull fresh postings from a list of employers you track, deduplicated across runs
• Recruiters — watch competitors and target accounts for new openings
• Market research — measure hiring by department across a sector over time
• Job seekers — monitor a shortlist of companies and see new roles the day they post
• AI agents — a clean, typed job feed to call as a tool

Notes and limits

• Public board endpoints only. No logins, no cookies, no access controls circumvented.
• Workday returns 20 postings per request and stops serving results past 2,000 per board.
• Workday's list response gives a location summary such as 5 Locations rather than the individual cities, and no department. That is what the board publishes.
• Descriptions are truncated to 5,000 characters.
• Recruitee, Personio, Teamtailor, and Comeet are not supported. Their endpoints could not be verified against a live board, and shipping a parser that has never seen real data is how scrapers end up silently returning nothing.

Pricing

Pay per event: you are charged per posting delivered. A scheduled run that finds nothing new delivers nothing and charges nothing.