Workday Jobs Scraper

Scrape jobs from any Workday-hosted careers site — those on *.myworkdayjobs.com (NVIDIA, Salesforce, Citi, Capital One, and thousands more). HTTP-only, no login, no cookies.

What it outputs (per job)

• type = job_workday
• id (requisition ID), requisitionId, jobPostingId, url, title
• jobDescription (HTML), descriptionText (stripped)
• location, primaryLocation (full descriptor + country block), additionalLocations, country, countryCode
• postedOn, postingDate (ISO), startDate, endDate, endDateText, applicationCloseDate, applicationTimeRemaining
• timeType, remoteType (when the tenant publishes it — e.g. Office - Flexible, Onsite, Remote)
• compensationRangeMin, compensationRangeMax, compensationCurrency, compensationFrequency (parsed best-effort from the description)
• canApply, hiringOrganization, jobRequisitionUrl
• questionnaireId, secondaryQuestionnaireId
• similarJobs (list of {title, url})
• logoImage (absolute URL, when the tenant exposes one)
• videoInfo (when present)
• tenant, site, externalPath
• scrapedAt

If the given URL has zero matching jobs (or Workday's global maintenance page is up for that tenant), a single job_workday_blocked sentinel record is emitted so runs exit 0.

Input

Field	Type	Description
startUrls	string[]	Workday tenant career URLs (e.g. https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite) or direct job URLs (contain /job/...).
searchText	string	Optional keyword filter (engineer, marketing, …).
locations	string[]	Optional list of Workday location UUIDs (facet values returned by the /jobs API under facets[].values[].id).
postedAt	string	Optional cutoff. Either ISO date (2026-04-01) or a relative string (7 days, 30 days).
startAt	string	Optional start-date cutoff, same format as postedAt.
scrapeDetails	boolean	Fetch the detail endpoint for full HTML descriptions + country + canApply. Default true.
maxItems	integer	Max jobs per run. Default 3, max 10000.
proxyConfiguration	object	Apify proxy. Datacenter is enough; Workday's public API is unauthenticated.

How it works

1. Parse tenant, region, site from each Workday URL.
2. POST the public API https://<tenant>.<region>.myworkdayjobs.com/wday/cxs/<tenant>/<site>/jobs with { limit, offset, searchText, appliedFacets } — paginated until maxItems or total is exhausted.
3. If scrapeDetails = true, GET each job's detail endpoint for the full jobPostingInfo (description HTML, country, canApply, startDate, timeType).
4. Apply postedAt / startAt cutoffs in-process.

FAQ

Does this work for every Workday tenant? Yes — every *.myworkdayjobs.com site exposes the same public /wday/cxs/ JSON API without auth.

What about bot blocking? Workday's public careers API does not challenge datacenter IPs. A small retry loop handles the occasional 429/503.

What's a location UUID? When you open a Workday careers site and expand the Locations filter, each entry has a stable UUID. The first API response's facets array returns them.

Why are some runs emitting only the sentinel? Individual tenants occasionally redirect to community.workday.com/maintenance-page during Workday's rolling deploys. The sentinel lets the caller retry later without a hard failure.