Workday Jobs Scraper — Any Workday Careers Site | $1/1K

Paste a Workday careers URL, get the open jobs as clean JSON. $1 per 1,000 jobs.

Large employers (NVIDIA, Adobe, Salesforce and thousands more) run their careers sites on Workday (<company>.wd5.myworkdayjobs.com/<site>). This Actor uses the public JSON API that the careers site itself calls to list the jobs, then reads each posting for its full description, every location, posting date and pay range. No browser and no proxy are needed.

• ✅ Any Workday careers site — paste the careers URL, a job URL, tenant.wdN/site, or just tenant/site (the Workday data centre — wd1, wd3, wd5, wd12… — is looked up for you)
• ✅ Server-side keyword search — keywords are sent to Workday's own search, so you can pull "engineer" jobs from a 2,000-job site without paging everything
• ✅ Complete postings — full description, all locations (Workday's "3 Locations" is expanded), employment type, remote flag, posting date
• ✅ Pay ranges parsed from the description text when the posting states one
• ✅ Monitor mode — pay only for jobs you have not received before
• ✅ Filters — keywords, excluded words, location, remote, posted within N days
• ✅ No login, no CAPTCHA handling, no personal data

Also want Greenhouse, Lever, Ashby and SmartRecruiters in the same run? Use the combined ATS Jobs API. This is the Workday-only version.

Use cases

• Enterprise hiring monitoring — track what large employers are hiring, by location and team
• Job boards and aggregators that need Workday coverage
• Alerts for new roles at specific companies (daily monitor run)
• Labor-market and salary research across big employers
• Recruiting-agency lead signals: which companies open many roles in a region

Input

{
    "careerSites": [
        "https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite",
        "adobe.wd5/external_experienced",
        "salesforce/External_Career_Site"
    ],
    "keywords": ["engineer"],
    "excludeKeywords": ["intern"],
    "locations": ["United States", "Remote"],
    "postedWithinDays": 14,
    "maxItemsPerSource": 200,
    "monitorMode": false
}

Field	What it does
careerSites	Required. One per line. Accepted: the careers URL (https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite, https://wd3.myworkdaysite.com/recruiting/<tenant>/<site>), a URL of any job on the site, nvidia.wd5/NVIDIAExternalCareerSite, or nvidia/NVIDIAExternalCareerSite. A company name alone is not enough — one company can have several Workday sites.
keywords	Keep jobs whose title contains any of these words. Sent to Workday's search, one query per keyword, results de-duplicated.
excludeKeywords	Drop jobs whose title contains any of these words.
locations	Keep jobs whose location, any extra location or country contains one of these words (whole-word, case-insensitive). remote matches remote jobs.
remote	any, remote_only or exclude_remote.
postedWithinDays	Only jobs posted in the last N days (Workday shows the date at day precision).
maxItemsPerSource / maxItems	Limits per site / per run (0 = no limit).
maxConcurrency	Detail requests in parallel (default 4, max 10). Requests to Workday are rate-limited to about 4 per second regardless.
monitorMode, monitorStateName	Only output jobs not delivered before; name of the store that remembers them.

How to find the careers URL

Open the company's careers page and click "Search for jobs". The address looks like https://<tenant>.wd5.myworkdayjobs.com/<site> — paste it as it is, locale segments such as /en-US/ are fine.

Output

{
    "job_id": "workday:nvidia:GenAI-Algorithms-Intern---Model-Optimization-for-Inference_JR2027377",
    "ats": "workday",
    "company": "Nvidia",
    "company_board_id": "nvidia",
    "title": "GenAI Algorithms Intern – Model Optimization for Inference",
    "department": null,
    "team": null,
    "location": "Poland, Warsaw",
    "locations": ["Poland, Warsaw", "Poland, Remote"],
    "country": "PL",
    "remote": true,
    "workplace_type": "remote",
    "employment_type": "FULL_TIME",
    "posted_at": "2026-10-20T00:00:00.000Z",
    "updated_at": null,
    "salary_min": 117750,
    "salary_max": 204100,
    "salary_currency": "PLN",
    "salary_period": "YEAR",
    "salary_source": "description",
    "description_text": "…",
    "apply_url": "https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite/job/Poland-Warsaw/GenAI-Algorithms-Intern---Model-Optimization-for-Inference_JR2027377/apply",
    "source_url": "https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite/job/Poland-Warsaw/GenAI-Algorithms-Intern---Model-Optimization-for-Inference_JR2027377",
    "input": "https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite",
    "scraped_at": "2026-10-20T16:50:26.671Z"
}

Workday-specific notes:

• Workday does not publish departments or teams; department and team are null (and there is no department filter).
• Salary is parsed from the posting text (salary_source: "description"); it is null when the text states no range.
• posted_at is Workday's posting date at day precision (00:00 UTC).
• workplace_type is remote when Workday or the title says so, otherwise null — Workday does not distinguish hybrid from onsite in its data.
• The company name is derived from the tenant (nvidia → Nvidia).

Speed and limits

• Workday needs one request per job for the description, so a run is about 1–2 jobs per second: 500 jobs take roughly 5 minutes. In our tests 162 NVIDIA internship jobs took about 100 seconds.
• One Workday search returns at most ~2,000 results. For very large sites add keywords (each keyword is a separate search) or use postedWithinDays plus monitor mode.
• Requests are limited to about 4 per second per host to stay polite.

Monitor mode

Turn on Monitor mode, save the input as a task and schedule it daily. The first run returns all matching jobs, later runs only jobs not delivered before (state is kept per site in the key-value store workday-jobs-monitor). Note that a monitor run still lists the whole site to find new jobs, so combine it with keywords or postedWithinDays on big sites. Tested: second run returned 0 jobs; after forgetting 3 jobs the third run returned exactly those 3.

Pricing

Event	Price
Job in the dataset	$0.001 ($1 per 1,000 jobs)
Actor start	$0.002 per run

500 jobs ≈ $0.50. Jobs removed by your filters and jobs already delivered in monitor mode are free. Use Maximum cost per run to cap spending.

FAQ

The site "was not found". Check the URL: it must be the careers site (…/<site>), not the company's marketing page. For tenant/site shorthand the Actor tries the common Workday data centres; if the site is not found, paste the full URL.

Why is the tenant/site shorthand slower? It sends one small request per data centre until it finds the site (about a second each at most).

Is it legal? The Actor reads public job postings from the same public endpoints the careers site uses, without login or any protection bypass, and outputs no personal data. You are responsible for how you use the data.