ATS With a Public Job Postings API: 33 Compared

Most popular applicant tracking systems expose a public JSON endpoint behind their hosted careers pages that returns a company's open jobs with no API key: Greenhouse, Lever, Ashby, SmartRecruiters, Workable, Recruitee, Teamtailor, BambooHR and Workday all work this way. A few, such as iCIMS, Taleo, Jobvite and Avature, have no public jobs API at all, so you collect job links from the career-site HTML and read each job page's schema.org JobPosting JSON-LD.

Everything below comes from the collectors we run against these 33 platforms, not from vendor marketing pages. If you'd rather not build and maintain 33 of them, CleanJobData normalizes all of these sources into one REST API: see the Jobs API docs.

Quick reference table

{company} and similar placeholders stand for the company's identifier on that platform (see how to find it). "Description in list?" says whether the list call already carries the full job description or you need one more request per job. In the salary column, "—" means we don't read a structured salary field from that source.

ATS with a public JSON API (no key)

These platforms answer a plain HTTP request with JSON. No account, no key, no browser session. They are the same endpoints the platforms' own careers pages call.

Greenhouse

One call returns every job on the board. Add content=true and each job carries its full HTML description, so there's no per-job request. Use first_published as the post date; updated_at moves whenever a recruiter edits the job. There's no salary field. Full walkthrough: Greenhouse job board API.

curl "https://boards-api.greenhouse.io/v1/boards/acme/jobs?content=true"

Lever

mode=json returns the whole board as a JSON array. The description is split across description, lists (the bulleted sections) and additional, so join all three. salaryRange has min, max, currency and an interval you need to annualize, and createdAt is a millisecond timestamp. See the Lever postings API guide.

curl "https://api.lever.co/v0/postings/acme?mode=json"

Ashby

Ashby's posting API returns all jobs with descriptionHtml in one response. Pass includeCompensation=true to get the compensation object, which has both a summary string and structured summaryComponents. The Ashby guide covers the pay and address objects.

curl "https://api.ashbyhq.com/posting-api/job-board/acme?includeCompensation=true"

SmartRecruiters

The postings list pages with limit and offset (100 per page) and returns summaries only. The job ad (jobAd.sections) needs one detail request per posting at /postings/{id}. There's no salary field. Details in the SmartRecruiters API guide.

curl "https://api.smartrecruiters.com/v1/companies/acme/postings?limit=100&offset=0"

Workable

The widget endpoint returns every published job, and details=true adds the full description to each one. Dates are in published_on; there's no salary field. See the Workable API guide.

curl "https://apply.workable.com/api/v1/widget/accounts/acme?details=true"

Recruitee

One call to /api/offers/ returns all open offers. The body is split between description and requirements, so concatenate them. The salary object has min, max, currency and period. More in the Recruitee guide.

curl -H "accept: application/json" "https://acme.recruitee.com/api/offers/"

Teamtailor

Every Teamtailor career site serves a JSON Feed at /jobs.json. Each item has the full content_html plus a schema.org _jobposting object; follow next_url until it's empty. The host is the career-site host, which can be a custom domain. See the Teamtailor jobs.json guide.

curl -H "accept: application/json" "https://acme.teamtailor.com/jobs.json"

BambooHR

/careers/list returns every opening in one call but without descriptions. Each job's description, datePosted and free-text compensation come from /careers/{id}/detail. See the BambooHR guide.

curl -H "accept: application/json" "https://acme.bamboohr.com/careers/list"

Workday

Workday's CXS API is a POST with a JSON body of limit, offset, appliedFacets and searchText, 20 jobs per page. The list gives a relative postedOn ("Posted 30+ Days Ago"); the detail call at /job/{externalPath} returns the full description and an exact startDate. See the Workday guide.

curl -X POST "https://acme.wd5.myworkdayjobs.com/wday/cxs/acme/External/jobs" \
  -H "content-type: application/json" \
  -d '{"appliedFacets":{},"limit":20,"offset":0,"searchText":""}'

Other platforms with public JSON

• Pinpoint. /postings.json returns every job with its description in one call. Salary is in compensation_minimum / compensation_maximum, but only use it when compensation_visible is true. The JSON has no posting date; the same site's /jobs.rss feed does.
• Rippling. The v2 list (page, pageSize=100) has structured locations but no description, date or pay. The v1 detail call (/platform/api/ats/v1/board/{slug}/jobs/{id}) has the description, createdOn and payRangeDetails.
• Zoho Recruit. /recruit/v2/public/Job_Openings needs a pagename parameter, which differs per company and is embedded in the career page as page_name. It returns the whole list in one call and rejects any extra query parameter such as page or limit. The description (Job_Description) comes from /Job_Openings/{id} with the same pagename.
• Eightfold. /api/apply/v2/jobs pages with start at a fixed 10 jobs per page, whatever you pass in num. The list's job_description is often missing, so use the detail call at /api/apply/v2/jobs/{id}. Some tenants answer the v2 list with a 403 and use a different search API (/api/pcsx/search) instead.
• PageUp. jobs.json returns every open job with the full Overview description in one array. Hosted tenants use /{tenantId}/{siteCode}/{locale}/jobs.json; white-labeled custom domains use /{locale}/listing/jobs.json. The two shapes aren't interchangeable.
• Oracle Recruiting Cloud. A REST endpoint under /hcmRestApi/resources/latest/ with paging inside the finder parameter (limit, offset). The description (ExternalDescriptionStr) comes from recruitingCEJobRequisitionDetails with a ById finder.
• ADP Workforce Now. OData-style $top / $skip paging on job-requisitions, but ADP returns about 20 rows per page whatever $top says, so advance by the number of rows you actually received. Descriptions come from /job-requisitions/{itemID}, and pay is in payGradeRange.
• UKG Pro Recruiting. LoadSearchResults is a POST with Top / Skip in the body and needs no cookie or token. The detail is an HTML page (OpportunityDetail?opportunityId={id}) with the job object embedded in an inline script, including structured locations and compensation fields.
• SAP SuccessFactors. Some Career Site Builder tenants answer a POST to /services/recruiting/v1/jobs with paged JSON. When they don't, sitemap.xml or the HTML /search?startrow=N page gives you the job URLs, and each job page carries schema.org microdata.
• Gem. The career page is a client-rendered shell that calls a public GraphQL endpoint. One JobBoardList query returns every posting; the description, publish time and compensationHtml come from a second ExternalJobPosting query, and both can be batched in one POST. Gem's documented Job Board API needs the employer's own key, so it can't be used for other companies' boards.

Public XML feeds: Personio and JazzHR

• Personio. https://{company}.jobs.personio.de/xml (or .jobs.personio.com) is a full XML export with descriptions in jobDescriptions and pay in salaryInformation. Request ?language=en first and fall back to the plain feed for positions whose English description is empty. Some accounts disable the feed; then you read the career page HTML and each job page. See the Personio XML feed guide.
• JazzHR. https://app.jazz.co/feeds/export/jobs/{company} has title, location, description and type for every open job. It doesn't carry the posting date; that comes from the JSON-LD on each job page at {company}.applytojob.com. Accounts can disable the export too.

curl -H "accept: application/xml" "https://acme.jobs.personio.de/xml?language=en"

ATS without a public jobs API

For these platforms there's no documented, keyless JSON endpoint. The data is still public, because it's on the career page, but you get it in one of three ways.

JSON tied to a career-page session

Cornerstone OnDemand, Paycom, Dayforce and IBM Kenexa BrassRing do serve jobs as JSON, but only to their own career pages: the search API expects the session the page sets up when it loads. These aren't public APIs, they change without notice, and each tenant can be configured differently, so treat them as internal endpoints rather than something to build a product on.

What you get once you're there differs: Dayforce and Paycom include pay rates on the job detail, Cornerstone's list returns a shortened description with the full ad on the detail, and BrassRing lets each employer name its own fields, so the title and location columns vary by tenant. If you need these platforms covered, this is where a maintained source like the CleanJobData Jobs API saves the most work.

Job data embedded in the career page

• Paylocity. The list page at recruiting.paylocity.com/recruiting/jobs/All/{guid} assigns every job to window.pageData in an inline script. Parse that JSON, then read the JSON-LD on each /Recruiting/Jobs/Details/{id} page for the description and salary.
• Comeet. Every Comeet page embeds COMPANY_POSITIONS_DATA, a JavaScript array of all open positions with full descriptions, so one HTML request gets everything.

HTML list plus JSON-LD on each job page

Here you collect job links from server-rendered HTML, then fetch each job page and parse its <script type="application/ld+json"> JobPosting block. That block uses the same schema.org fields Google for Jobs reads: title, description, datePosted, jobLocation, baseSalary. See our JobPosting schema guide for the field reference.

• iCIMS. Page through /jobs/search?ss=1&in_iframe=1 with a zero-based pr parameter, then read each /jobs/{id}/job?in_iframe=1.
• Taleo. searchResults pages by rowFrom in steps of 25; job pages are viewRequisition?org={org}&cws={cws}&rid={rid}. Some tenants have no usable JSON-LD and need HTML parsing.
• Jobvite. jobs.jobvite.com/{company}/jobs lists every job link; each /job/{id} page has the JSON-LD.
• Breezy HR. The career-site root lists /p/{id}-{slug} links; each job page has the JSON-LD, including baseSalary when the employer sets one.
• Freshteam. /jobs renders every job in one page. Job pages have JSON-LD, but the description is HTML-entity-encoded inside the JSON string and remote is the string "true" or "false".
• Avature. SearchJobs pages by jobOffset, 10 links per page. Job pages are plain HTML label/value pairs that differ by tenant, so the description has to be picked out of them. Many tenants publish sitemap.xml with a per-job lastmod.

How to find a company's ATS and identifier

Start from the company's careers page and follow the "Apply" or "View jobs" link. The host usually names the platform, and the path or subdomain is the identifier:

Many companies embed the board on their own domain instead, so the URL bar shows careers.acme.com. In that case open your browser's developer tools, switch to the Network tab, filter by Fetch/XHR, and reload the careers page. The request that returns the job list shows the platform's API host and the identifier in its URL or body. This is also how you find the values that never appear in the address bar, such as Zoho's pagename or Cornerstone's regional API host.

There's no public directory that maps every company to its ATS identifier. Finding one board you already know about is easy; doing it for thousands of employers is a separate problem.

Things that differ between platforms

Salary availability.

• Structured pay fields exist on Lever (salaryRange), Ashby (compensation), Recruitee (salary), Pinpoint, Rippling, Personio, ADP, UKG and Dayforce, and wherever JSON-LD carries baseSalary (iCIMS, Jobvite, Taleo, Breezy, Paylocity, Paycom).
• Greenhouse, Workable, SmartRecruiters, Teamtailor, Workday and Oracle give you no salary field we read, so pay only shows up in description text.
• Units vary: hourly, monthly and yearly figures sit in the same field, marked by Lever's interval, Recruitee's period, Personio's type or JSON-LD's unitText. Annualize before you compare. More in the salary extraction guide.
• A salary field isn't always a number. PageUp's Salary is usually placeholder prose like "Commensurate", BambooHR's compensation is free text, and Pinpoint only means it when compensation_visible is true.

Location formats.

• Some sources are structured: Comeet and Cornerstone give ISO alpha-2 countries, Rippling v2 and Zoho give separate city, state and country fields, and ADP's shortName reads " City, ST, CC".
• Others are free text: Greenhouse's location.name, PageUp's LocationList (sometimes a campus name on both sides of the pipe), and Personio office names.
• Structured doesn't mean correct. Freshteam's JSON-LD swaps addressRegion and addressLocality, and Comeet's state field was wrong on real postings we checked.
• JazzHR and Zoho give full country names ("United States"), so you need a name-to-code lookup. See the location normalization guide.

Pagination.

• Don't trust the page size you asked for. ADP ignores $top, Eightfold ignores num, Dayforce is fixed at 25 and BrassRing at 50. Advance by the rows you actually received and stop on an empty page or the reported total.
• Offset paging over a sort with many ties can skip or repeat jobs. On BrassRing, sorting by the default date or title lost jobs across pages; sort by a unique key when the API allows it, and dedupe by job ID either way.
• Workday can report a total larger than it will actually page through, so stop on an empty page as well.

Published-date reliability.

• Prefer the field that records first publication: Greenhouse first_published over updated_at, Gem firstPublishedTsSec, Eightfold t_create.
• Some lists have no date at all: Pinpoint's JSON (use the RSS feed), Rippling's v2 list and Gem's list (use the detail call), and JazzHR's XML (use the job page).
• Some dates are coarse or oddly encoded: Workday's list gives relative text, Dayforce's list date has no real time, Zoho's Date_Opened is date-only, Cornerstone uses M/D/YYYY, and PageUp uses .NET /Date(1791295200000)/ strings.
• Some platforms only expose a last-updated date: BrassRing's lastupdated and Comeet's time_updated are the only dates available.

Detecting closed jobs. None of these platforms send webhooks to outsiders. Poll the list, diff job IDs, and treat a job as closed when it drops out of a complete list. A partial list (a failed page, a truncated response) should never expire jobs. See syncing job data in depth.

FAQ

Which ATS have a public API for job postings?

Greenhouse, Lever, Ashby, SmartRecruiters, Workable, Recruitee, Teamtailor, BambooHR, Pinpoint, Rippling, Zoho Recruit, Eightfold, PageUp, Workday, Oracle Recruiting Cloud, ADP Workforce Now, UKG Pro Recruiting and Gem all serve job listings as JSON with no API key. Personio and JazzHR publish keyless XML feeds. iCIMS, Taleo, Jobvite, Avature, Breezy HR and Freshteam have no public jobs API, and Cornerstone, Paycom, Dayforce and BrassRing need a token or session from the career page first.

Do I need an API key to get Greenhouse jobs?

No. https://boards-api.greenhouse.io/v1/boards/{board_token}/jobs?content=true is public and returns every published job with full descriptions. Greenhouse's Harvest API does need a key, but that's for the employer's own internal hiring data, not their public job board.

Is scraping ATS job boards allowed?

These endpoints serve job listings that employers publish so the public can find and apply to them, and none of the ones above require you to log in. That doesn't settle every question. Each platform has its own terms of service, employers may have their own terms on their career sites, and some data uses carry extra obligations. Check the relevant terms for your use case, respect robots.txt, keep your request rate modest, cache what you've fetched, and identify your client honestly. This isn't legal advice; if it matters for your business, ask a lawyer.

How do I get jobs from every ATS at once?

You either build and maintain a collector per platform, using the table above, or use an API that already does it. CleanJobData collects from these ATS platforms and serves the jobs in one normalized schema: see the Jobs API docs.

Does Workday have a public jobs API?

Yes, though it's undocumented. Every Workday career site calls POST https://{tenant}.wd{N}.myworkdayjobs.com/wday/cxs/{tenant}/{site}/jobs with no key, 20 jobs per page, and you need one detail call per job for the description and exact date. The Workday guide has the request body and response shape.