Workday, Greenhouse, Lever & Ashby Job Scraper + New-Job Alerts

Track open roles at any company that hires through Workday, Greenhouse, Lever, Ashby, Workable, SmartRecruiters or Recruitee, in one clean, normalized dataset. Paste a careers URL or just the company website (e.g. nvidia.com) and the Actor finds the job board for you. Schedule it and every run returns only the jobs that are new (or changed) since the last run, so you pay for fresh signal instead of re-downloading the same postings.

• ✅ 7 ATSs, 1 schema: title, department, locations, remote/hybrid/onsite, employment type, salary range, posted date, apply link and description.
• ✅ Workday career sites (*.myworkdayjobs.com): the ATS behind most Fortune 500 careers pages. Complete job lists even on boards with more than 2,000 openings, where Workday's own search stops at 2,000.
• ✅ Domain → job board discovery: enter stripe.com, nvidia.com or figma.com; the Actor reads the company's careers page, finds its board on any of the 7 ATSs, and remembers it.
• ✅ Monitor mode: remembers what it has already seen, per company, between scheduled runs.
• ✅ Official public job-board feeds only: the same JSON feeds the companies' own careers pages use. No login, no proxies, no browser, no blocking.
• ✅ Privacy-safe: recruiter contact details are never returned, and e-mail addresses are stripped from descriptions.
• ✅ Fast: about 11,000 Workday jobs from 12 companies in about 75 s, 7,400 jobs from 13 Greenhouse/Lever/Ashby/Workable/SmartRecruiters/Recruitee boards in about 60 s, and repeat monitor runs in under a minute.

Who uses it

• Sales and recruiting agencies: get a daily list of companies that just opened roles you can fill or sell into ("hiring signals").
• Job seekers and job boards: build niche boards ("remote Rust jobs at YC companies") that refresh themselves.
• Investors and analysts: watch hiring velocity by company, department and location as a growth indicator.
• HR and compensation teams: benchmark salary ranges disclosed under pay-transparency laws.

How to use

1. Add companies to Companies / job boards, one per line. Any of these work:

   • https://boards.greenhouse.io/stripe or https://job-boards.greenhouse.io/stripe
   • https://jobs.lever.co/spotify (EU boards on jobs.eu.lever.co work too)
   • https://jobs.ashbyhq.com/linear
   • https://apply.workable.com/huggingface
   • https://jobs.smartrecruiters.com/BoschGroup
   • https://bunq.recruitee.com
   • https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite (any Workday careers URL, including /en-US/... and job links, or wdN.myworkdaysite.com/recruiting/...)
   • greenhouse:stripe, lever:spotify, workday:nvidia.wd5/NVIDIAExternalCareerSite, and so on
   • stripe.com, https://www.figma.com/careers or Hugging Face: the Actor finds which ATS the company uses (see Auto-discovery)

2. Pick a Mode:

   • New and changed jobs (monitor) returns jobs not seen before plus jobs whose title, location, salary or description changed. This is the default for API calls that leave mode out.
   • New jobs only ignores edits.
   • All open jobs returns a full snapshot every run. The Console form opens on this mode so a first test run always shows results.

   Scheduling? Switch the mode to New and changed jobs or New jobs only before you save the schedule or task. Otherwise every run returns (and bills) the full list of open jobs.

3. Optional filters: title keywords (include and exclude), location keywords, department, remote only, workplace type, and posted within N days.

4. Schedule it (e.g. daily) in Apify Console. Connect the dataset to Slack, e-mail, Google Sheets, Zapier, Make or a webhook to get alerts.

First run: every open job counts as new. To start monitoring without paying for the existing backlog, enable First run: record baseline only. Several watch lists? Give each schedule its own Monitor ID.

Output

Each item is one job:

{
  "jobId": "ashby:openai:fc5bbc77-a30c-4e7a-9acc-8a2e748545b4",
  "changeType": "new",
  "provider": "ashby",
  "companyToken": "openai",
  "companyName": "openai",
  "title": "Data Engineer",
  "department": "Applied AI",
  "team": "Applied AI Engineering",
  "locations": ["San Francisco", "Mountain View"],
  "locationText": "San Francisco",
  "country": "United States",
  "isRemote": false,
  "workplaceType": "hybrid",
  "employmentType": "FULL_TIME",
  "salaryMin": 235000,
  "salaryMax": 385000,
  "salaryCurrency": "USD",
  "salaryPeriod": "YEAR",
  "salarySource": "structured",
  "postedAt": "2023-09-26T22:34:48.884000+00:00",
  "jobUrl": "https://jobs.ashbyhq.com/openai/fc5bbc77-a30c-4e7a-9acc-8a2e748545b4",
  "applyUrl": "https://jobs.ashbyhq.com/openai/fc5bbc77-a30c-4e7a-9acc-8a2e748545b4/application",
  "descriptionHtml": "<h3><strong>About the team</strong></h3>...",
  "descriptionText": "ABOUT THE TEAM\n\nThe Applied team works across research...",
  "firstSeenAt": "2026-10-19T23:48:09+00:00",
  "scrapedAt": "2026-10-19T23:48:09+00:00",
  "monitorId": "default"
}

Export as JSON, CSV, Excel, XML or RSS, or read it through the Apify API.

A run summary is saved to the run's key-value store as OUTPUT: companies checked, success rate, and per-company counts for new, changed, unchanged and removed jobs. Companies that failed are listed in ERRORS. Neither record is billed.

Field notes

Field	Notes
salary*	salarySource: "structured" comes from the ATS's own pay field (Ashby, Lever, Recruitee). "parsed" is extracted from pay-transparency text in the description, e.g. "$150,000 – $200,000" or "€55.000—€65.000 EUR". When a posting lists several regional ranges, the first one is used.
companyName	Lever, Ashby and Workday don't publish a display name, so the board token (Workday: the tenant, e.g. nvidia) is used.
jobId (Workday)	workday:TENANT.wdN/SITE:REQ_ID, built from the requisition ID at the end of the job URL, so it stays stable when a title is edited.
employmentType	Normalized to FULL_TIME, PART_TIME, CONTRACT, TEMPORARY, INTERN or OTHER. Greenhouse rarely exposes it.
descriptionHtml / descriptionText	SmartRecruiters and Workday need one extra request per job for the description. It is fetched only for jobs being returned, up to SmartRecruiters description fetch cap / Workday detail fetch cap (default 300) per company.
Workday fields	The Workday job list gives title, location, posting age and link. The detail request (see above) adds the description, exact posting date, employment type (full/part time), remote/hybrid/onsite where the employer sets it, country code and every location of multi-location jobs. Workday doesn't publish a department on its public feed. Jobs above the detail cap keep the list fields; their postedAt comes from "Posted N days ago" (null for "30+ days ago").
Workday counts	The run summary shows boardReportedTotal (the "N jobs found" number on the careers page) next to jobsFetched. A few boards also count postings that have no title or link; those are reported as boardUnlinkedPostings and not returned.
changeType	new or changed. unchanged appears only in All open jobs mode.

Auto-discovery

Enter a company website or domain (stripe.com, https://www.figma.com/careers) or a company name (Hugging Face):

1. Careers page scan (websites/domains). The Actor opens the home page, follows the site's own "Careers" / "Jobs" links (plus /careers, /jobs and careers. subdomains), and looks for links, embeds and inline data pointing at any of the 7 ATSs, Workday included. At most 6 pages per company are read, one at a time.
2. Slug probe. If no board link is found (many careers pages load jobs with JavaScript), likely board names are tried on Greenhouse, Lever, Ashby, Workable, Recruitee and SmartRecruiters (datadoghq.com → datadog). Where the ATS reports a company name, it must match your input, which guards against slug collisions; for example, example.com is not matched to an unrelated "Democorp" board. Workday URLs can't be guessed, so Workday boards are found through step 1 or by pasting the careers URL.

The run summary shows how each company was found (discoveryMethod: careers-page or slug-probe). In our test on 29 real company domains, 25 resolved to the right board (86%); the misses were careers sites that hide the ATS entirely (paste the careers URL for those). A successful discovery is billed once as Company auto-discovered and then cached, so scheduled runs aren't billed for it again. Board URLs and provider:token inputs are never billed for discovery. If two inputs point at the same board (e.g. stripe.com and greenhouse:stripe), its jobs are returned and billed once.

Pricing (pay per event)

Event	Price
Job returned (each job written to the dataset)	$9 per 1,000 jobs on the Free and Bronze plans, $4.50 on Silver, $3 on Gold and higher
Company auto-discovered (name/domain input only, once per company)	$0.005
Actor start	Apify's standard start fee

In monitor mode you pay only for new or changed jobs. Example: watching 200 companies daily that post about 300 new roles a day across them costs about $2.70 per day ($0.90 on Gold). The best-known career-site job feed on the Store charges $12 per 1,000 jobs. Set Max cost per run in Console to cap spending. When the cap (or Max jobs per run) is reached, the Actor stops cleanly, and anything not returned is delivered on the next run.

Reliability

• Retries with exponential backoff on timeouts, 429s and 5xx errors, honouring Retry-After.
• Per-ATS concurrency limits, so no board is hammered. Workday boards are read with at most 3 parallel requests each, spaced out; company websites are read one page at a time, at most 6 pages per company.
• One company failing never fails the run. It is reported in ERRORS and its saved state is left untouched, so nothing is missed next time.
• State is written per company only after its jobs are stored. An interrupted or capped run never "loses" jobs.

Legal and data

The Actor reads only the official public job-board APIs that each ATS provides for publishing openings (Greenhouse Job Board API, Lever Postings API, Ashby Posting API, Workable jobs widget, SmartRecruiters Posting API, Recruitee Careers Site API, and the JSON feed behind Workday careers sites). It doesn't log in, bypass access controls or collect personal data. Recruiter names and e-mail addresses are dropped. You are responsible for using the data in line with applicable laws and each provider's terms.

FAQ

Can it monitor LinkedIn or Indeed? No. It covers company-owned ATS boards only. That's where most jobs are posted first, and it's reliable and permitted.

How do I find a company's board? Open its careers page. The apply links usually point to myworkdayjobs.com, greenhouse.io, lever.co, ashbyhq.com, workable.com, smartrecruiters.com or recruitee.com. Or just enter the company website, e.g. nvidia.com.

Does it get every job on huge Workday boards? Yes. Workday's search returns at most 2,000 jobs per query, so on bigger boards the Actor reads them category by category and removes duplicates. In testing, NVIDIA's board showed "2,000" but has 2,762 jobs, and all 2,762 were returned.

Why did a company return 0 jobs? It may have no open roles, or it may have moved to another ATS. Check OUTPUT.perCompany.

Something broke or you need another ATS? Open an issue on the Actor's Issues tab.