Workday Jobs Scraper

Get published job postings from any company that hires through Workday (career sites on myworkdayjobs.com or myworkdaysite.com) as clean, consistent JSON. Thousands of large employers use Workday, including NVIDIA, Salesforce, Adobe, Intel and Walmart.

The Actor reads the same public job data that the company's career page shows to every visitor. There is no login, no proxies and no CAPTCHA handling, and it never submits applications.

Use it to track hiring at target companies, feed a job board, build lead lists from hiring signals, or watch for new roles with onlyNew.

Job descriptions belong to the employers who published them.

Input

{
  "careerSites": [
    "https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite",
    "https://salesforce.wd12.myworkdayjobs.com/External_Career_Site"
  ],
  "searchText": "data engineer",
  "keywords": [],
  "locations": ["Germany", "Remote"],
  "remoteOnly": false,
  "postedSinceDays": 7,
  "includeDetails": true,
  "onlyNew": false,
  "maxJobs": 200
}

• careerSites: up to 30 Workday career site URLs, copied from the browser. Language parts such as /en-US/ and links to a single job on that site both work.
• searchText: passed to the career site's own search box.
• keywords, locations: case-insensitive "contains" filters on the title and on location or country.
• remoteOnly: keep only jobs the employer marks as remote.
• postedSinceDays: keep only jobs published within this many days. Older jobs are skipped before their pages are opened, so this also makes runs faster.
• includeDetails: open each job for the full description, exact posting date, country and remote type (default on). Turn it off for a quick list of titles and locations.
• onlyNew: return only jobs you have not received before for the same query. Use stateKey to share this memory between queries.
• maxJobs: maximum number of jobs per run (1–10,000). It is shared evenly between the career sites, so every company gets its turn.
• maxJobsPerSite: optional limit per career site, if you want a different split.

Output

One dataset item per job:

{
  "source": "workday",
  "company": "nvidia",
  "companyName": "NVIDIA",
  "careerSite": "https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite",
  "jobId": "JR2001234",
  "title": "Senior Data Engineer",
  "department": null,
  "team": null,
  "locations": ["US, CA, Santa Clara", "US, TX, Austin"],
  "country": "United States of America",
  "remote": null,
  "employmentType": "Full time",
  "compensation": null,
  "postedAt": "2026-10-25T00:00:00Z",
  "postedOnText": "Posted 2 Days Ago",
  "updatedAt": null,
  "applyUrl": "https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite/job/...",
  "jobUrl": "https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite/job/...",
  "descriptionText": "…",
  "descriptionHtml": null,
  "scrapedAt": "2026-10-27T12:00:00Z"
}

The field names match our sister Actors for Greenhouse, Lever and Ashby and for SmartRecruiters, Workable and Personio, so results can be merged directly.

A run summary (jobs listed per site, sites not found, errors) is saved to the OUTPUT record of the run's key-value store.

Good to know

• With includeDetails off, Workday shows only "2 Locations" for jobs in several places, so locations is empty for those jobs.
• Workday shows a posting date only as "Posted 3 Days Ago". With includeDetails on, the exact date is taken from the job page; without it, the date is calculated from that text, and jobs marked "30+ Days Ago" get postedAt: null.
• remote is true or false only when the employer publishes a remote type or writes "Remote" in the location; otherwise it is null.
• Workday does not publish salary as a structured field, so compensation is always null. Pay ranges, when given, are inside the description text.
• Workday lists at most about 2,000 jobs per search. For very large employers, use searchText or several narrower runs.
• A site that does not exist is reported in notFound and does not fail the run.

Pricing

You pay only for jobs returned.