Greenhouse, Lever & Ashby Jobs Scraper API — New & Closed Job Monitor

By Veinwork. A Greenhouse jobs scraper, Lever jobs scraper and Ashby jobs scraper in one Actor: get lightweight, structured job postings and hiring signals directly from company career boards (ATS job boards), ready for AI agents, spreadsheets and CRMs. Run it once for a jobs export, or schedule it as a job monitor that reports new and closed jobs. No login or API key needed.

1. Why use ATS Jobs Monitor?

• $0.50 per 1,000 delivered jobs. Closed-job notices and filtered-out jobs are free. A small start event also applies.
• Small records for AI agents and tables. Titles, departments, locations, remote status, salary numbers, dates and links without bulky descriptions. Our November 1, 2026 sample of 1,365 jobs averaged 0.75 KB per record; individual records vary. Open postingUrl for the full description.
• Monitor hiring signals. onlyNew delivers newly seen jobs with free closed-job notices. Each run writes a per-company SUMMARY with open, matching, new, delivered and closed counts for sales signals and investment research.
• Transparent collection. Official public APIs only, Lever crawl-delay respected, unlisted Ashby jobs excluded, and an honest User-Agent.

2. Input and examples

Field	Default	Meaning
companies	Required	1–100 board URLs, ats:slug entries or bare slugs for automatic detection. Prefixes: greenhouse, lever, ashby. Duplicate ATS/slug pairs are processed once. Entries that cannot be parsed are skipped and reported with status invalid in SUMMARY and the run warnings; the run fails only when no entry is usable.
titleKeywords	[]	Keep titles containing any keyword; up to 50 strings, case-insensitive substring matching.
excludeTitleKeywords	[]	Drop titles containing any keyword; up to 50 strings, case-insensitive.
locationKeywords	[]	Keep jobs with any listed location containing any keyword; up to 50 strings, case-insensitive.
remoteOnly	false	Keep only jobs identified as remote.
postedWithinDays	Unset	Optional integer 1–365; includes the cutoff. Jobs without a publication date are excluded when enabled.
onlyNew	false	Deliver newly seen jobs, plus free closure notices. State is saved under STATE in the named key-value store ats-jobs-monitor-state.
maxJobsPerCompany	1000	Maximum matching jobs delivered per company per run; integer 1–10000.

One-time collection from three companies:

{
  "companies": ["airbnb", "https://jobs.lever.co/zoox", "ashby:ramp"]
}

Remote engineering jobs:

{
  "companies": ["greenhouse:airbnb", "lever:outreach", "ashby:ramp"],
  "titleKeywords": ["engineer"],
  "excludeTitleKeywords": ["intern"],
  "remoteOnly": true
}

Daily monitoring input (configure a daily Apify Schedule separately):

{
  "companies": ["greenhouse:airbnb", "lever:outreach", "ashby:ramp"],
  "onlyNew": true,
  "maxJobsPerCompany": 1000
}

The first successful check treats all open jobs as new. Later checks compare against the named store. New jobs withheld by the per-company cap or cost limit remain new next run. Failed companies keep their previous state and do not generate false closure notices.

3. Output

The default dataset contains free closed rows first, then delivered job rows. Both use the same field set. Dates are UTC; missing values are null; currencies and salary intervals retain the source's meaning. isNew is included even when onlyNew is false. Illustrative examples follow.

A job row:

{
  "recordType": "job",
  "jobKey": "greenhouse:airbnb:8241468",
  "ats": "greenhouse",
  "companySlug": "airbnb",
  "companyName": "Airbnb",
  "jobId": "8241468",
  "title": "Senior Demand Generation Specialist, Hotels (12M FTC)",
  "department": "Program Management",
  "team": null,
  "location": "France, Italy",
  "allLocations": [
    "France, Italy"
  ],
  "remote": null,
  "workplaceType": null,
  "employmentType": null,
  "salaryMin": 74000.0,
  "salaryMax": 92000.0,
  "salaryCurrency": "EUR",
  "salaryInterval": null,
  "postedAt": "2026-09-30T12:12:47Z",
  "updatedAt": "2026-10-12T18:02:19Z",
  "postingUrl": "https://careers.airbnb.com/positions/8241468?gh_jid=8241468",
  "applyUrl": "https://careers.airbnb.com/positions/8241468?gh_jid=8241468",
  "isNew": true,
  "closedDetectedAt": null
}

A closed-job row (illustrative):

{
  "recordType": "closed",
  "jobKey": "greenhouse:airbnb:previous-role",
  "ats": "greenhouse",
  "companySlug": "airbnb",
  "companyName": null,
  "jobId": "previous-role",
  "title": "Previous engineering role",
  "department": null,
  "team": null,
  "location": null,
  "allLocations": [],
  "remote": null,
  "workplaceType": null,
  "employmentType": null,
  "salaryMin": null,
  "salaryMax": null,
  "salaryCurrency": null,
  "salaryInterval": null,
  "postedAt": null,
  "updatedAt": null,
  "postingUrl": null,
  "applyUrl": null,
  "isNew": false,
  "closedDetectedAt": "2026-11-01T12:00:00Z"
}

One company record in the SUMMARY array in the default key-value store:

{
  "input": "greenhouse:airbnb",
  "ats": "greenhouse",
  "companySlug": "airbnb",
  "companyName": "Airbnb",
  "status": "ok",
  "openJobs": 3,
  "matchedJobs": 3,
  "newJobs": 2,
  "deliveredJobs": 2,
  "closedJobs": 1,
  "error": null
}

openJobs counts unique jobs before filters. matchedJobs counts matching jobs before the cap; newJobs counts newly seen matching jobs; deliveredJobs reflects actual delivery after the cost limit. closedJobs counts eligible closure notices. Output links expose the dataset overview and SUMMARY. Summary status values are ok, not_found, error, and invalid. Company errors appear in the summary and warnings; other companies can still succeed.

4. Pricing and cost limits

Event	Price	Charged for
job	$0.0005 per job ($0.50 per 1,000)	Each delivered job row
apify-actor-start	$0.0005 per GB-memory start event	A run start; default 512 MB counts as one event

Closed-job notices, summaries and filtered-out jobs are not charged. A zero-new-job run can still incur the start event. Under pay-per-event pricing, delivered job rows equal job events charged.

When the configured cost limit prevents further delivery, the run stops pushing jobs and reports stopped at the cost limit. Free closure notices are delivered first, even with zero job budget. Undelivered new jobs remain eligible next run. Reaching the limit exactly after the final job is complete delivery, not a truncated run.

5. Sources and responsible collection

Only official public APIs are used:

• Greenhouse: https://boards-api.greenhouse.io/v1/boards/{slug}/jobs?pay_transparency=true, with departments from https://boards-api.greenhouse.io/v1/boards/{slug}/departments.
• Lever: https://api.lever.co/v0/postings/{slug}?mode=json.
• Ashby: https://api.ashbyhq.com/posting-api/job-board/{slug}?includeCompensation=true.

No full job descriptions or employer-written long text are exported. Open postingUrl for the original. Lever's crawl-delay is respected: requests, including retries and detection probes, start at least one second apart. Greenhouse embed URLs may be parsed as input but are never requested. Ashby jobs with isListed: false are excluded; compensation is exported only when explicitly marked for display.

Requests identify themselves as VeinworkATSJobsMonitor/0.1 (+https://apify.com/veinwork/ats-jobs-monitor). No login, authentication headers, rotating proxies or browser impersonation are used. If a provider requires authentication or blocks access, the service is withdrawn rather than bypassing the restriction.

6. Known limitations

• Only Greenhouse, Lever and Ashby US-region API hosts are supported. EU hosts job-boards.eu.greenhouse.io and jobs.eu.lever.co are not supported. This host restriction does not exclude jobs located outside the US.
• This is a company-board monitor, not a global job database or web-wide keyword search engine.
• Greenhouse does not provide team, employment type or salary interval in this output. Lever and Ashby do not provide company display names. Missing fields remain null.
• Jobs missing an ID, title or posting link are omitted. Descriptions and salary narrative text are not included.
• Newness is tracked per company, independently of filters: filtered-out new jobs still count as seen. Changing filters does not make those jobs new again. Runs using the same named store share history; avoid overlapping monitoring runs that could overwrite it.
• Closure notices use title filters only; location, remote and date filters cannot be applied to a stored historical title.
• Many Lever companies make runs slower because requests are spaced by at least one second. Detection and retries can add requests and time.
• A run accepts at most 100 company inputs. Detection tries Greenhouse, Lever, then Ashby; exhausted request errors stop detection for that company instead of guessing.

7. FAQ

How do I find the slug?

Copy a supported board URL such as https://jobs.lever.co/zoox. Its first path segment, zoox, is the slug. You can enter lever:zoox directly. A bare slug such as airbnb enables automatic detection; an explicit prefix avoids extra probes. Whitespace in company names is removed and the result is lowercased for automatic detection; for example, Open AI is detected as openai.

How do I run this daily?

Save monitoring input as an Apify task and attach a daily Schedule. Set onlyNew to true and retain the same state store. Starting this Actor once does not create a recurring schedule.

How do I connect a webhook?

Configure an Apify integration or webhook for a successful run. Your receiver can retrieve jobs and closures from the run's default dataset and company counts from the default key-value store's SUMMARY. Check status and summary errors before treating a successful run as complete coverage.