ATS Jobs Scraper: Greenhouse, Lever, Ashby, Workable & More avatar

ATS Jobs Scraper: Greenhouse, Lever, Ashby, Workable & More

Pricing

$2.00 / 1,000 jobs

Go to Apify Store
ATS Jobs Scraper: Greenhouse, Lever, Ashby, Workable & More

ATS Jobs Scraper: Greenhouse, Lever, Ashby, Workable & More

Paste company career-page URLs and get every open job in one JSON format, whichever applicant tracking system the company uses: Greenhouse, Lever, Ashby, Workable, Recruitee, Teamtailor and more. $0.002 per job ($2 per 1,000), no start fee. Not affiliated with any ATS vendor.

Pricing

$2.00 / 1,000 jobs

Rating

0.0

(0)

Developer

fecklesspleb

fecklesspleb

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

1

Monthly active users

7 hours ago

Last modified

Features

Limited permissions

Share

ATS Jobs Scraper API: jobs from nine applicant tracking systems in one schema

ATS Jobs Scraper is a multi-ATS job scraper and jobs API: it reads company career pages on nine applicant tracking systems (Greenhouse, Lever, Ashby, Workable, Recruitee, Workday and Teamtailor, plus Polymer and HireHive when you give their board URL or a polymer: / hirehive: prefix) and outputs every open job in one consistent JSON schema: title, department, locations, remote/hybrid/onsite status, seniority, posting and expiry dates, salary (also annualized) when published, apply link, a cross-board dedupe key and full description. The same run can optionally also read the public Remote OK and Jobicy feeds (remoteBoards). It costs $0.002 per job written to the dataset ($2 per 1,000 jobs), with no start fee. It is an independent scraper and is not affiliated with or endorsed by any of these ATS vendors or job boards.

Sample output: the first item from a local run of the prefilled example input ({"companies": ["https://boards.greenhouse.io/airbnb"], "maxTotalJobs": 4}) on 2 November 2026, shortened to 16 of its 42 fields (descriptionText and the rest are in the full item):

{
"id": "8306554",
"ats": "greenhouse",
"companySlug": "airbnb",
"companyName": "Airbnb",
"title": "AirCover Specialist, Claims Experience",
"department": "Community Support",
"location": "Canada",
"workArrangement": "remote",
"postedAt": "2026-10-28T14:52:05+00:00",
"salaryMin": 58000,
"salaryMax": 68000,
"salaryCurrency": "CAD",
"salaryAnnualUsdMin": 42340,
"salaryAnnualUsdMax": 49640,
"dedupeKey": "e6075f853f93e2173ffb",
"isAgency": false
}

Paste board URLs, ats:slug pairs, bare slugs (the Actor works out which ATS the company uses; Polymer and HireHive need a board URL or a polymer: / hirehive: prefix) or careers pages that embed a board. Or search a built-in directory of 694 verified Greenhouse, Ashby and Lever boards. Optionally set remoteBoards to also read Remote OK and/or Jobicy in the same run. Filter by keyword, location, department and date, and turn on only-new mode (onlyNewJobs) on scheduled runs to get only jobs you haven't seen before. No ATS login or API key is needed.

At a glance

  • One row per open job on any supported ATS: title, company, location (plus the full location list), remote flag, posting date, job URL and apply link.
  • Also: department, work arrangement, seniority, employment type, salary range (with annual and USD-converted figures) when the job states one, and the full description (text, HTML or both), in one schema for every supported ATS.
  • A Jobs table in the Output tab with company, title, location, posting date and URL, and export to JSON, CSV or Excel.
  • A stable uid per job for upserts, and only-new mode (onlyNewJobs) that returns only jobs you haven't seen before.
  • Optional public remote-job feeds (remoteBoards): Remote OK and Jobicy, in the same schema. The default is none, so runs that omit it are unchanged. Each remote-board job is charged the same $0.002 job event as an ATS job.
  • $0.002 per job, nothing for jobs your filters remove. The input form opens with a sample input (Airbnb's Greenhouse board, 4 jobs, $0.008), so you can press Start and see real rows before adding your own boards.

Who it's for

  • Job board and job-feed builders who need one feed across many companies, whatever ATS each one uses.
  • Recruiters and sourcers watching hiring at a list of target companies.
  • Sales, RevOps and ABM teams using new job postings as hiring signals for account lists.
  • Investors and analysts tracking headcount plans and hiring trends across a portfolio or market.
  • Job seekers who want a daily digest of new roles at the companies they care about.

What data you get

One dataset item per job, in the same format for every ATS. Jobs from the optional remote boards use that schema too, with the extra feed fields below. Our single-ATS Actors use the same fields.

FieldWhat it holds
idThe job's ID in the ATS (on Workday, normally the requisition ID).
atsOn an ATS job: greenhouse, lever, ashby, workable, recruitee, workday, teamtailor, polymer or hirehive. On a remote-board job: the board name.
companySlugThe board slug (on Workday, tenant/site, e.g. utaustin/UTstaff). null on jobs from the optional remote boards.
companyNameThe company name published by the board or its API (for example Palantir Technologies on Lever, Ramp on Ashby). If the board publishes none and you entered a company name (not a slug or URL), the name as you typed it. Otherwise null. On Workday, which has no display-name field, it is the hiring entity without its leading company code, e.g. Example Corp for 2100 Example Corp; on UT Austin's site it is The University of Texas at Austin (null if the detail page could not be read). It is never the slug.
titleJob title.
department, teamAs published by the ATS (null when it has none).
location, locations, countryMain location, all listed locations, and country when the ATS gives one.
isRemote, workplaceType, workplaceTypeRawRemote flag; the ATS's workplace value normalized to remote, hybrid, onsite or null (null when the ATS says nothing clear, such as Workday's Flexible); and the ATS's own text (e.g. on-site on Lever, OnSite on Ashby). On Greenhouse these come from the company's custom fields such as Workplace Type or Employment Type, and are null when it publishes none.
workArrangementNormalized: remote, hybrid, onsite or null. Taken from the ATS's workplace field first, then from "remote"/"hybrid"/"on-site" in the title or location.
employmentType, employmentTypeRawNormalized to the schema.org values FULL_TIME, PART_TIME, CONTRACTOR, TEMPORARY, INTERN, VOLUNTEER, PER_DIEM, OTHER (a value that fits none of them, such as Permanent) or null; and the ATS's own text, e.g. Full-time (Lever), FullTime (Ashby), Full time (Workday).
postedAt, updatedAtWhen the ATS exposes them: an ISO 8601 timestamp on most ATSs, a calendar date (YYYY-MM-DD, no time) on Workday.
postedAtPrecisionexact (a date from the ATS), day or approximate (worked out from a label such as "Posted 3 Days Ago"), atLeast ("Posted 30+ Days Ago": the job is at least that old), or null.
expiresAtClosing date when the ATS publishes one (Teamtailor's validThrough), otherwise null.
url, applyUrlJob page and application link.
salaryMin, salaryMax, salaryCurrency, salaryPeriod, salaryTextPay range when present.
salaryAnnualMin, salaryAnnualMaxThe pay range as a yearly amount in the posting's own currency: hour × 2,080, day × 260, week × 52, month × 12. No currency conversion. Null when there is no range or no pay period.
salaryAnnualUsdMin, salaryAnnualUsdMaxThe annual range in whole US dollars, converted with a fixed table of rough reference rates bundled with the Actor (not live rates). Null when there is no salary, no pay period, or the currency is not in the table.
salarySourceats when the ATS has a pay field, description when the range was parsed from the description text, otherwise null.
hiringEntity, requisitionIdWorkday: legal hiring entity and requisition ID. Workable: the job code. Greenhouse: the company's requisition field when it holds an ID-like code; placeholders (ONE, See Opening ID) and free-text notes become null. Null where the ATS has none.
descriptionText / descriptionHtmlFull description in the format you choose.
seniorityintern, junior, mid, senior, lead, staff, principal, manager, director, vp, executive or null, read from the job title with word rules. A hint, not a verified level: manager means the title says manager, which includes roles such as Product Manager.
isAgencytrue when the job looks like it was posted by a staffing or recruitment agency: the company name (Randstad, Hays, Robert Half, "… Staffing", "… Recruitment", "Personalberatung" and similar) or client wording in the description ("on behalf of our client", "für unseren Kunden suchen wir"). A heuristic: an in-house job at an agency is also true, and an agency with a neutral name and no client wording is false. With includeDescription off only the company name is checked, except on Workday, where the description is always read.
uidStable key ats:slug:id on ATS jobs, handy for upserts. On a remote-board job it is source:id (the same uid Remote Jobs Aggregator gives that job).
dedupeKeyHash of company, title and first location after removing case, accents and gender suffixes such as (m/f/d). The same job on two boards, or re-posted with a new ID, gets the same key.
crossSourceKeyHash of company (legal form such as Inc or GmbH removed) and title, with no location, normalised like dedupeKey. The same job on an ATS board and a remote board, or on two ATSs, gets the same key even when the locations are written differently. null when the company or title is missing. The Workday Jobs Scraper and Remote Jobs Aggregator output the same key for the same job.
source, sourceName, attributionOn jobs from the optional remote boards (kept from the feed): the board identifier, the source name, and the credit line the source asks for. If you display the job, keep this text and link back to the job URL.
category, tags, companyLogoOn jobs from the optional remote boards, kept from the feed.
firstSeenAt, lastSeenAtWhen a run first and last returned the job. Remembered across runs when trackFirstSeen or onlyNewJobs is on; otherwise both are this run's time.
scrapedAtWhen the run read the job.

Sample item from Ashby (real output; long text shortened for this page):

{
"id": "34413f8d-26bf-4bbc-8ade-eb309a0e2245",
"ats": "ashby",
"companySlug": "ramp",
"companyName": "Ramp",
"title": "Security Engineer, Cloud",
"department": "Engineering",
"team": "Backend",
"location": "New York, NY (HQ)",
"locations": [
"New York, NY (HQ)",
"Remote (Canada)",
"Remote (US)",
"Miami, FL"
],
"country": "USA",
"isRemote": true,
"workplaceType": "hybrid",
"employmentType": "FULL_TIME",
"workplaceTypeRaw": "Hybrid",
"employmentTypeRaw": "FullTime",
"postedAt": "2026-04-07T17:12:35.753000+00:00",
"updatedAt": null,
"url": "https://jobs.ashbyhq.com/ramp/34413f8d-26bf-4bbc-8ade-eb309a0e2245",
"applyUrl": "https://jobs.ashbyhq.com/ramp/34413f8d-26bf-4bbc-8ade-eb309a0e2245/application",
"salaryMin": 211400,
"salaryMax": 290600,
"salaryCurrency": "USD",
"salaryPeriod": "YEAR",
"salaryText": "$211.4K – $290.6K • Offers Equity",
"salarySource": "ats",
"hiringEntity": null,
"requisitionId": null,
"uid": "ashby:ramp:34413f8d-26bf-4bbc-8ade-eb309a0e2245",
"scrapedAt": "2026-10-29T01:10:47.004320+00:00",
"expiresAt": null,
"postedAtPrecision": "exact",
"salaryAnnualMin": 211400,
"salaryAnnualMax": 290600,
"salaryAnnualUsdMin": 211400,
"salaryAnnualUsdMax": 290600,
"isAgency": false,
"workArrangement": "hybrid",
"seniority": null,
"dedupeKey": "0d8ecfae9515641f1178",
"firstSeenAt": "2026-10-29T01:10:47.004320+00:00",
"lastSeenAt": "2026-10-29T01:10:47.004320+00:00"
}

Sample item from Workday (UT Austin UTstaff). The board fields, uid, dedupeKey and crossSourceKey are real output from workday-jobs-scraper 1.1.1 smoke run ilpMFiKmQhVUMdH3B on 10 November 2026. The fields this Actor adds (expiresAt, postedAtPrecision, salaryAnnual*, workArrangement, seniority, firstSeenAt, lastSeenAt, isAgency) were computed by running this Actor's own code on that item and its description, at the run's time. descriptionText removed. The fields and their order match this Actor's Workday output with the default input:

{
"id": "R_00049214",
"ats": "workday",
"companySlug": "utaustin/UTstaff",
"companyName": "The University of Texas at Austin",
"title": "Manager, Clinical Research Operations",
"department": null,
"team": null,
"location": "AUSTIN, TX",
"locations": [
"AUSTIN, TX"
],
"country": "United States of America",
"isRemote": false,
"workplaceType": null,
"employmentType": "FULL_TIME",
"postedAt": "2026-11-10",
"updatedAt": null,
"url": "https://utaustin.wd1.myworkdayjobs.com/UTstaff/job/AUSTIN-TX/Manager--Clinical-Research-Financial-Operations_R_00049214",
"applyUrl": "https://utaustin.wd1.myworkdayjobs.com/UTstaff/job/AUSTIN-TX/Manager--Clinical-Research-Financial-Operations_R_00049214/apply",
"salaryMin": null,
"salaryMax": null,
"salaryCurrency": null,
"salaryPeriod": null,
"salaryText": null,
"salarySource": null,
"hiringEntity": "The University of Texas at Austin",
"requisitionId": "R_00049214",
"workplaceTypeRaw": null,
"employmentTypeRaw": "Full time",
"uid": "workday:utaustin:R_00049214",
"expiresAt": null,
"postedAtPrecision": "day",
"salaryAnnualMin": null,
"salaryAnnualMax": null,
"salaryAnnualUsdMin": null,
"salaryAnnualUsdMax": null,
"workArrangement": null,
"seniority": "manager",
"dedupeKey": "0d9a8451fdb712901e48",
"crossSourceKey": "3135273bab7ffaceddc3",
"firstSeenAt": "2026-11-10T19:46:32.267129+00:00",
"lastSeenAt": "2026-11-10T19:46:32.267129+00:00",
"isAgency": false,
"scrapedAt": "2026-11-10T19:46:32.267129+00:00"
}

(descriptionText removed from both samples to keep the page short.)

How far to trust the derived fields

  • USD conversion is approximate. salaryAnnualUsdMin / salaryAnnualUsdMax use fixed exchange rates bundled with the Actor, set in October 2026 (RATES_AS_OF in my_actor/fx.py). They are rough reference values, not live exchange rates, and they are not refreshed at run time, so treat the USD figures as salary bands, not exact amounts.
  • isAgency, visaSponsorship and yearsExperienceMin / yearsExperienceMax are heuristics. They come from company-name and description patterns, not from a field the employer sets, and can be wrong in both directions. Check the posting before relying on them.
  • The visa and years filters are off by default and slow runs down. visaSponsorship (default any) and maxYearsExperience (default empty) read each job's description, so while either is on the Actor fetches job details for every job, even with includeDescription off. Expect more requests and longer runs.
  • Invalid input fails at the start. A negative number, a minimum salary above the maximum, or an unknown option value stops the run before any request with a status message that lists every problem. Nothing is fetched or charged.

How to use it

Supported ATSs and input formats

ATSPasteAuto-detected from a bare slugIn the directory
Greenhousehttps://boards.greenhouse.io/airbnb, greenhouse:airbnbYes147 boards
Lever (US and EU)https://jobs.lever.co/spotify, lever:palantirYes58 boards
Ashbyhttps://jobs.ashbyhq.com/ramp, ashby:rampYes489 boards
Workablehttps://apply.workable.com/<company>YesNo
Recruiteehttps://<company>.recruitee.comYesNo
Workdayhttps://utaustin.wd1.myworkdayjobs.com/UTstaffNo: paste the full career site URLNo
Teamtailorhttps://proxima.teamtailor.com, teamtailor:proximaYesNo
Polymerhttps://jobs.polymer.co/<company>, polymer:<company>No: paste the board URL or polymer: slugNo
HireHivehttps://<company>.hirehive.com, hirehive:<company>No: paste the board URL or hirehive: slugNo
  • Bare slugs or company names such as stripe, Stripe or Hugging Face: the Actor first looks the name up in the built-in directory, then tries slug spellings of it (huggingface, hugging-face, without suffixes such as Inc or GmbH, at most four spellings) on Greenhouse, Lever, Ashby, Workable, Recruitee and Teamtailor. It uses the board with the most open jobs for the first spelling that finds one, and the log says which slug and ATS it picked.
  • Entries that don't resolve (no board found, an unsupported ATS or a careers page with no board link) each get one dataset record with status: "unresolved", the input as written, a reason, and the triedSlugs and triedAts. These records are free: they are not charged as jobs.
  • Careers page URLs such as https://www.notion.com/careers: the page is read and any embedded or linked board is used. This only works when the page actually links to or embeds a supported board. If it doesn't, the entry is skipped with a message.
  • Not supported: SmartRecruiters (its API's robots.txt disallows automated access); Personio (its API terms restrict using the job feed to build public directories); Breezy HR, BambooHR, Jobvite and Rippling (their terms of use do not allow automated copying of their sites); and ATSs not in the table above.

Running it

  1. Add companies to Companies (career board URLs or slugs), one per line. Any mix of the formats above works, up to 2,000 per run.
  2. Or turn on Search the built-in company directory to add 694 verified Greenhouse, Ashby and Lever boards. Max directory companies keeps only the N boards that had the most jobs when the directory was built (0 = all).
  3. Optionally choose Remote job boards (remoteBoards) to also read Remote OK and/or Jobicy. Max jobs per company caps the jobs per remote board. A run with remote boards and no companies is a normal run, not the demo.
  4. Add filters so you only pay for jobs you want.
  5. Click Start, then export from the Output tab as JSON, CSV or Excel.

With no companies and the directory off (for example an empty {} input), the run is a demo: it returns up to 4 sample jobs from https://boards.greenhouse.io/airbnb, charged at the normal per-job price (at most 4 job events), and stops after 50 seconds at the latest. The log says Demo mode: no boards/companies given, returning up to 4 sample jobs from https://boards.greenhouse.io/airbnb. The input form is prefilled with that board and Max jobs in total = 4. That sample input (only the demo board, with a job cap of 1 to 4 in Max jobs in total or Max jobs per company) runs as the same demo, so it also costs at most $0.008; the log then says Demo mode: prefilled sample input, returning up to 4 sample jobs from https://boards.greenhouse.io/airbnb. Replace the board, or clear or raise the cap, for a normal run.

Input fieldLabelNotes
companiesCompanies (career board URLs or slugs)Board URLs, ats:slug, bare slugs or careers page URLs.
searchDirectorySearch the built-in company directoryDefault false.
maxDirectoryCompaniesMax directory companies0 = all (default).
remoteBoardsRemote job boards (optional)Default empty (none). Also read Remote OK (remoteok) and Jobicy (jobicy) in the same run. Each job is charged the same job event as an ATS job. Jobs have uid source:id (as in Remote Jobs Aggregator), companySlug: null, source, sourceName and an attribution text; link back to the job URL when you display them. Jobicy is read at most once per hour (its fair-use rule): choosing jobicy in remoteBoards opens the named key-value state store (stateStoreName, default ats-jobs-scraper-state) even when only-new options are off, and records the last pass under remote-jobicyLastPass; a run within an hour of the last pass skips it (status: "deferred" in RUN_SUMMARY). With remote boards and no companies, the run is a normal run, not the demo. Max jobs per company caps jobs per remote board.
jobicyGeo, jobicyIndustry, jobicyTagRemote board optionsSame meaning as in Remote Jobs Aggregator. Only used with remoteBoards including jobicy. jobicyTag is passed unchanged to Jobicy's API as tag. Jobicy's API docs describe tag as a keyword search: the API page says it is applied to the job title; the OpenAPI spec (2.2.19) says across available job content and accepts 3–50 characters (ASCII). This Actor does not check the value or filter on it.
keywordsTitle keywords (any)Keep jobs whose title contains at least one word (case-insensitive).
excludeKeywordsExclude title keywordsDrop jobs whose title contains any of these words, e.g. intern.
searchInDescriptionMatch keywords in description tooAlso match keywords against the description. Needs includeDescription on every ATS except Workday, where the detail page is always read.
locationsLocations (any)Keep jobs whose location contains one of these strings, e.g. London, United States, Remote.
remoteOnlyRemote jobs onlyKeep jobs marked remote by the ATS or with "remote" in the location.
departmentsDepartments or teams (any)Keep jobs whose department or team contains one of these strings. Workday jobs have no department, so this filter removes all Workday jobs.
postedWithinDaysPosted within last N days1–3650. Empty keeps all.
includeUndatedJobsKeep jobs without a posting dateDefault true. Applies when postedWithinDays is set.
hasSalaryOnly jobs with a salaryDefault false. Keeps only jobs that state a salary (any currency or period).
minAnnualSalaryUsdMinimum annual salary (USD)Keeps jobs whose salaryAnnualUsdMax (or only figure) is at least this. Jobs without a convertible salary are dropped unless includeJobsWithoutSalary is on.
maxAnnualSalaryUsdMaximum annual salary (USD)Keeps jobs whose salaryAnnualUsdMin (or only figure) is at most this.
includeJobsWithoutSalaryKeep jobs without a salaryDefault false. Under a USD bound, also keeps jobs whose salary is missing or cannot be converted. Dropped jobs are never charged; RUN_SUMMARY gets salaryJobsExcluded.
timeRangePosted within (fixed window)Optional: 1h, 24h, 7d, 30d or all. Replaces postedWithinDays when set; all turns the date limit off.
workArrangementWork arrangementremote / hybrid / onsite, derived from the ATS field, location and title. Jobs with no detectable arrangement are left out when set.
employmentTypesEmployment typeFULL_TIME, PART_TIME, CONTRACTOR, TEMPORARY, INTERN, PER_DIEM, VOLUNTEER, OTHER. Jobs with no type are left out when set.
senioritySeniorityintern … executive, derived from the title. Jobs whose title shows no level are left out when set.
visaSponsorshipVisa sponsorshipany (default, off), yes (description says sponsorship is offered) or not_no (drop jobs that say no sponsorship). Pattern-based, not a legal statement. Adds visaSponsorship / visaSponsorshipEvidence; fetches job details.
maxYearsExperienceMax years of experienceDrops jobs whose description asks for more years; jobs stating no number are kept. Adds yearsExperienceMin / yearsExperienceMax; fetches job details.
excludeAgenciesExclude recruiting agenciesDefault false. Leaves out jobs with isAgency: true. They are dropped before the dataset push, so they are never charged; RUN_SUMMARY gets agencyJobsExcluded and each board agencyExcluded. On Workday, maxJobsPerCompany is then counted after this filter, so the site is paged further.
dedupeByRemove duplicate jobsDefault none. job drops a job with the same company, title and first location as one already output in this run. companyTitle merges a company's jobs with the same title into one record (locations holds all their locations; duplicateCount and mergedUids are added). Duplicates are removed before the dataset push, so they are never charged; RUN_SUMMARY gets duplicatesDropped. Only duplicates within one run are handled.
onlyNewJobsOnly new jobsReturn only jobs not returned by an earlier run with this option on. Default false. Needs a usable state store (stateStoreName); if it cannot be opened, every matching job is returned and charged on each run.
newOnlyByOnly new jobs: what counts as seenDefault uid (as before): a job is skipped when its uid was output by an earlier run. uidOrCrossKey: a job is also skipped when an earlier run output a job with the same crossSourceKey, so a re-post with a new ID, or the same job on another board, is not returned (or charged) again. Only used with onlyNewJobs. Keys are kept for 120 days under SEEN_XKEY in the state store, saved from the first run with this option on. A company that posts one title in several cities gets one key for all of them, so only the first is returned. RUN_SUMMARY gets skippedSeenCrossKey.
stateStoreNameState store nameNamed key-value store for only-new state. Use one name per scheduled feed. Default ats-jobs-scraper-state.
includeExpiredReport expired jobsDefault false. On scheduled runs, adds a free record with status: "expired" for each job an earlier run returned that its board no longer lists. Only boards read completely in this run are checked. While on, job records get status: "active". Uses the state store.
trackFirstSeenTrack first-seen datesRemember firstSeenAt across runs without dropping jobs you already got. Default false. Use a different stateStoreName from any onlyNewJobs feed.
redactContactsRemove contact details from descriptionsDefault true. Replaces personal e-mail addresses, phone numbers and "Your recruiter: name" lines in descriptions. Shared mailboxes such as jobs@ or accommodations@ are kept.
includeDescriptionInclude descriptionDefault true. Workday detail pages are always read, so locations, requisition ID and company come back filled in. This setting only controls whether the description text is included.
descriptionFormatDescription formattext (default), html or both.
maxJobsPerCompanyMax jobs per company0 = no limit (default).
maxTotalJobsMax jobs in total0 = no limit (default). The run also stops at your maximum cost per run. The form is prefilled with 4 for the sample run.
maxConcurrencyCompanies in parallel1–20, default 5. Requests to each host are rate limited regardless.

Example 1: new data and ML jobs at four specific companies, for a daily schedule.

{
"companies": [
"https://boards.greenhouse.io/airbnb",
"ashby:ramp",
"https://jobs.lever.co/spotify",
"https://utaustin.wd1.myworkdayjobs.com/UTstaff"
],
"keywords": [
"data engineer",
"machine learning"
],
"postedWithinDays": 7,
"onlyNewJobs": true
}

Example 2: search the directory for product designer roles in the last 14 days, capped at 500 jobs.

{
"searchDirectory": true,
"keywords": [
"product designer"
],
"locations": [
"Remote",
"New York"
],
"postedWithinDays": 14,
"maxTotalJobs": 500
}

Example 3: optional remote boards only (Remote OK and Jobicy). This is an input shape, not a measured run. Max jobs per company caps jobs per remote board.

{
"remoteBoards": [
"remoteok",
"jobicy"
],
"maxJobsPerCompany": 5
}

Input examples

Each of these was run locally on November 1, 2026; the job counts are what those runs returned.

Example 1: One company. This is the prefilled sample input, so it runs as the demo described above: at most 4 jobs, $0.008. It returned 4 jobs.

{
"companies": [
"https://boards.greenhouse.io/airbnb"
],
"maxTotalJobs": 4
}

Example 2: Several companies on different ATSs, up to 2 jobs from each. It returned 6 jobs ($0.012).

{
"companies": [
"https://boards.greenhouse.io/airbnb",
"ashby:ramp",
"https://jobs.lever.co/spotify"
],
"maxJobsPerCompany": 2
}

Example 3: Filter by title keyword and location (engineer jobs that are remote or in the United States, at most 5 in total). It returned 5 jobs ($0.010).

{
"companies": [
"https://boards.greenhouse.io/airbnb",
"ashby:ramp",
"https://jobs.lever.co/spotify"
],
"keywords": [
"engineer"
],
"locations": [
"Remote",
"United States"
],
"maxTotalJobs": 5
}

Set maxTotalJobs and maxJobsPerCompany to 0 (no limit) to get every matching job.

Output of example 1: the first item from that run, with descriptionText shortened for this page.

{
"id": "8306554",
"ats": "greenhouse",
"companySlug": "airbnb",
"companyName": "Airbnb",
"title": "AirCover Specialist, Claims Experience",
"department": "Community Support",
"team": null,
"location": "Canada",
"locations": [
"Canada"
],
"country": null,
"isRemote": true,
"workplaceType": "remote",
"employmentType": null,
"postedAt": "2026-10-28T14:52:05+00:00",
"updatedAt": "2026-10-28T14:52:05+00:00",
"url": "https://careers.airbnb.com/positions/8306554?gh_jid=8306554",
"applyUrl": "https://careers.airbnb.com/positions/8306554?gh_jid=8306554",
"salaryMin": 58000,
"salaryMax": 68000,
"salaryCurrency": "CAD",
"salaryPeriod": "YEAR",
"salaryText": "$58,000—$68,000 CAD",
"salarySource": "description",
"hiringEntity": null,
"requisitionId": null,
"descriptionText": "Airbnb was born in 2007 when two hosts welcomed three guests to their San Francisco home, and has since grown to over 5 million hosts who have …",
"workplaceTypeRaw": "Remote",
"employmentTypeRaw": null,
"uid": "greenhouse:airbnb:8306554",
"expiresAt": null,
"postedAtPrecision": "exact",
"salaryAnnualMin": 58000,
"salaryAnnualMax": 68000,
"salaryAnnualUsdMin": 42340,
"salaryAnnualUsdMax": 49640,
"workArrangement": "remote",
"seniority": null,
"dedupeKey": "e6075f853f93e2173ffb",
"firstSeenAt": "2026-11-01T14:37:44.637515+00:00",
"lastSeenAt": "2026-11-01T14:37:44.637515+00:00",
"isAgency": false,
"scrapedAt": "2026-11-01T14:37:44.637515+00:00"
}

After each run, the default key-value store holds RUN_SUMMARY: jobs pushed, unresolvedRecords, HTTP requests, stoppedByCostLimit, and one entry per board with found, matched, pushed, skippedSeen and status. status is ok, partial (a Workday site where some jobs past the 2,000-row cap could not be reached; partialReason says why), skipped, unresolved or error. A Workday entry also gets pathlessSkipped when the site listed jobs with no job page; those are skipped and not charged.

Expired jobs (includeExpired): each run remembers the jobs it returned per board (key <board>-live in the state store). When a later run reads the whole board and a remembered job is no longer listed, it adds one record with status: "expired", uid, id, title, url, companyName, lastSeenAt and expiredDetectedAt. These records are not charged (no job event), and the Expired jobs dataset view lists them. A board counts as read completely only when nothing cut the read short: not maxJobsPerCompany, maxTotalJobs, the maximum cost per run or an abort, and on Workday not rows past the 2,000-row cap that facet splits could not reach. Jobs hidden by your filters still count as listed, so changing filters never reports jobs as expired. A job that comes back after it expired is returned as new. RUN_SUMMARY gets expiredRecords, and each board entry gets expired and complete.

Workday notes

  • Workday sites must be pasted as the full career site URL (https://<tenant>.<wdN>.myworkdayjobs.com/<site>). https://wdN.myworkdaysite.com/recruiting/<tenant>/<site> works too and is read through the tenant's myworkdayjobs.com address, whose robots.txt applies. A careers page that links to or embeds a Workday site is also recognized. Locale segments such as /en-US are ignored.
  • Before reading a site, the Actor checks the site path (/<site>/, for example /InternalJobs/) against the tenant's robots.txt. If it is disallowed, the site is skipped and the log and RUN_SUMMARY say so. Internal or invite-only sites are usually disallowed this way.
  • Title keyword filters are applied to the Workday job list before the detail pages are fetched, so filtered-out jobs don't cost a detail request.
  • Every Workday job's detail page is read, whatever includeDescription is set to, so employmentType, country, requisitionId, hiringEntity, companyName and the full location list are filled in either way. Turning includeDescription off only leaves the description out of the output; it does not save requests.
  • Workday shows posting dates as labels such as "Posted Today" or "Posted 3 Days Ago". postedAt is the date that label implies, counted on the US Eastern calendar, with postedAtPrecision day. For "Posted 30+ Days Ago" the posting start date from the detail page is used and the precision is exact. If the detail page can't be read, the date is worked out from the label alone and the precision is atLeast.

Teamtailor notes

  • Teamtailor jobs come from the board's public RSS feed. Pay appears only when the company publishes it on the job, and expiresAt only when the job has a closing date.

Polymer notes

  • Polymer boards are read through Polymer's unauthenticated Public API for job board integrations (api.polymer.co/v1/hire). Polymer is only used when you paste a jobs.polymer.co board URL or a polymer:<company> entry; bare slugs and company names are never tried on Polymer.
  • With includeDescription on, each job's detail is read (one extra request per job) for the description and department. With it off, only the job list pages are read.
  • Pay comes from the board's salary line (for example "260K - 300K USD a year") when the company shows one; the original text is kept in salaryText.

HireHive notes

  • HireHive boards are read through HireHive's public Jobs API (https://<company>.hirehive.com/api/v2/jobs), which needs no key or login. HireHive is only used when you paste a <company>.hirehive.com URL or a hirehive:<company> entry; bare slugs and company names are never tried on HireHive.
  • The job list already carries the description, so includeDescription costs no extra requests. Pay comes from the job's compensation tier (minimum, maximum, currency and interval) when the company publishes one; the posted pay text is kept in salaryText.
  • An unknown company (HireHive answers 404) is reported as an unresolved entry.

Remote job boards

  • Opt in with remoteBoards: remoteok (Remote OK) and/or jobicy (Jobicy). The default is empty (none), so runs without it are unchanged.
  • jobicyGeo, jobicyIndustry and jobicyTag have the same meaning as in Remote Jobs Aggregator and apply only when remoteBoards includes jobicy. Each is passed unchanged to Jobicy's API (geo, industry, tag); an empty value is not sent.
  • Jobicy is read at most once per hour (its fair-use rule). Choosing jobicy opens the named key-value state store (stateStoreName, default ats-jobs-scraper-state) even when only-new options are off, and records the last pass under remote-jobicyLastPass. A run within an hour of the last Jobicy pass skips Jobicy (status: "deferred" in RUN_SUMMARY); if nothing else is read, one free status item is written.
  • Jobicy is asked for 100 jobs per request (count=100), the limit on Jobicy's API page; its OpenAPI spec (2.2.19) allows up to 200. More jobs come from Jobicy's cursor pages (same filters, up to 50 requests), so a maxTotalJobs or maxJobsPerCompany above 100 is still reached when Jobicy has enough matching jobs. At most 5,000 Jobicy jobs are read per run.
  • If a saved input still lists weworkremotely, that board is not read (its API terms forbid building a job search service). The run ignores it, writes one free status item saying so, and does not fail. A run that only names that board is not turned into the demo.
  • Remote-board jobs use uid source:id, companySlug is null, and they include source, sourceName and an attribution text kept from the feed. If you display these jobs, credit the source board and link back to the job URL. Do not use a board's logo without permission.
  • Each remote-board job is charged the same job event as an ATS job. With remote boards and no companies, the run is a normal run, not the demo.
  • With Max jobs in total set and both remote boards chosen, the two boards take turns: one job from each in remoteBoards order, each board in its own feed order, so a board that answers later is not left out. A job dropped by a filter or by dedupeBy does not use a turn; when one board runs out (or reaches Max jobs per company), the other fills the rest. A board waits for the other's turn, so a slow board can make the run take a little longer. Jobs from companies boards do not take part in the turns.

Chain with other jobs Actors

This Actor is one of four jobs Actors that share one job format and one set of keys (Workday Jobs Scraper, Career Site Jobs Feed, Remote Jobs Aggregator). You can pass the dataset of a run of any of them in datasetId; this run then skips the jobs that are already there. All four inputs below are optional and off by default: without them a run behaves exactly as before.

InputDefaultWhat it does
datasetIdemptyID of a dataset from an earlier run (this Actor or another jobs Actor). Its job items are read before any board. For an Apify integration, put {{resource.defaultDatasetId}} here. When empty and the run was started by an Apify integration, the integration payload's resource.defaultDatasetId is tried.
datasetModemergemerge: the earlier jobs are copied into this run's dataset first (free, no job event), then the new jobs follow, so the dataset is the combined, deduplicated list. exclude: nothing is copied; the dataset has only the new jobs.
datasetMatchByuiduid: a job counts as already there when the same posting (same uid) is in the earlier dataset. uidOrCrossKey: also when the same company and title (crossSourceKey) is there from any source; this can merge one title posted in several cities.
suiteOutputfalsetrue: every job item has all fields of the shared format in the shared order (missing ones null or []). Values are not changed. Jobs copied from datasetId are kept as they were.
  • A job that is already in the earlier dataset is dropped before it is pushed, so it is never charged. Only new jobs are charged (one job event each, same price as today) and only new jobs count toward maxTotalJobs.
  • Status items and expired or unresolved records in the earlier dataset are not copied. At most 100,000 items are read.
  • datasetId asks Apify for read-only access to that one dataset, so chaining works with this Actor's limited permissions; the run gets no access to your other storages. Only a dataset given in datasetId is granted this way, so the payload fallback may not be readable; give the dataset in datasetId.
  • If the dataset cannot be read, or datasetMode or datasetMatchBy has an unknown value, the run fails with an input error before any board is read (as for other invalid input).
  • onlyNewJobs and its state (ats-jobs-scraper-state) work as before and are applied too.
  • The run summary (RUN_SUMMARY in the key-value store) gets a previousDataset block with the counts read, copied and skipped.
  • Known limit: with datasetId and no companies and no remoteBoards, the run still uses the Airbnb demo board, as an empty input does today.

Pricing

This Actor uses pay-per-event pricing: $0.002 per job written to the dataset. That is $2.00 per 1,000 jobs. There is no separate start fee in this Actor's pricing (as listed on October 28, 2026; the Pricing tab on the Store page is the source of truth).

You are charged only for jobs that end up in your dataset. Jobs removed by your filters, jobs skipped in only-new mode and duplicates are not charged.

ExampleJobsCost
Five companies, filtered to one team150$0.30
Directory search with filters500$1.00
Large scheduled run (illustration)10,000$20.00

The last row is an illustration, not a measured run size. Use maxTotalJobs or a maximum cost per run for a hard ceiling.

Worked example: 1,000 jobs × $0.002 = $2.00.

Jobs from the optional remote boards (remoteBoards) are charged the same job event as ATS jobs. For remote boards only, Remote Jobs Aggregator is the cheaper remote-only option ($0.001 per job versus $0.002 in this Actor).

Sample run: the prefilled input and an empty {} input both return at most 4 jobs, so they cost at most 4 × $0.002 = $0.008.

To cap spend, set Maximum cost per run in the run options (or maxTotalChargeUsd via the API). The Actor stops cleanly when the next job would go over the limit and records stoppedByCostLimit: true in the run summary.

If you only need one ATS, our single-ATS Actors for Greenhouse, Lever and Ashby cost $0.001 per job with the same fields.

Integrations and API

The actor ID is fecklesspleb/ats-jobs-scraper (in API URLs: fecklesspleb~ats-jobs-scraper).

n8n, Make and Zapier. Each has an Apify integration with a "run Actor" action. Pick fecklesspleb/ats-jobs-scraper, paste the input JSON shown below, then read the results from the run's default dataset in the next step. To process only new jobs on a schedule, turn on Only new jobs and run it from a schedule (Apify Schedules or a trigger in your automation tool).

Webhooks and schedules. Apify Schedules can run the Actor on a cron, and a webhook on "run succeeded" can call your own endpoint with the run ID.

HTTP API. Run the Actor and get the items back in one call. This waits for the run to finish, so use it for small runs:

APIFY_API="https://api.apify.com/v2/actors"
curl -X POST "$APIFY_API/fecklesspleb~ats-jobs-scraper/run-sync-get-dataset-items?token=$APIFY_TOKEN" \
-H "Content-Type: application/json" \
-d '{"companies": ["https://boards.greenhouse.io/airbnb", "ashby:ramp", "https://jobs.lever.co/spotify", "https://utaustin.wd1.myworkdayjobs.com/UTstaff"], "keywords": ["data engineer", "machine learning"], "postedWithinDays": 7, "onlyNewJobs": true}'

Python (pip install apify-client):

from decimal import Decimal
from apify_client import ApifyClient
client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor("fecklesspleb/ats-jobs-scraper").call(
run_input={
"companies": [
"https://boards.greenhouse.io/airbnb",
"ashby:ramp",
"https://jobs.lever.co/spotify",
"https://utaustin.wd1.myworkdayjobs.com/UTstaff"
],
"keywords": [
"data engineer",
"machine learning"
],
"postedWithinDays": 7,
"onlyNewJobs": True
},
max_total_charge_usd=Decimal("5"), # optional spend cap
)
# apify-client 3.x returns a Run object; on 1.x/2.x use run["defaultDatasetId"]
for item in client.dataset(run.default_dataset_id).iterate_items():
print(item)

JavaScript / Node.js (npm install apify-client):

import { ApifyClient } from 'apify-client';
const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('fecklesspleb/ats-jobs-scraper').call(
{
"companies": [
"https://boards.greenhouse.io/airbnb",
"ashby:ramp",
"https://jobs.lever.co/spotify",
"https://utaustin.wd1.myworkdayjobs.com/UTstaff"
],
"keywords": [
"data engineer",
"machine learning"
],
"postedWithinDays": 7,
"onlyNewJobs": true
},
{ maxTotalChargeUsd: 5 }, // optional spend cap
);
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);

Results can also be exported from the dataset as JSON, CSV, Excel, XML or RSS from the Apify Console or the dataset API.

Responsible use

The Actor reads only public career pages and job endpoints, logged out, and only where the site's robots.txt allows it. Entries that robots.txt blocks are skipped and noted in the log. If a host's robots.txt can't be read (a server error, a page that isn't a robots file, or a network failure), that host is skipped with a warning in the log, on every ATS. A missing robots.txt (4xx) means no rules. Some employers' own website terms restrict automated access, so you are responsible for checking the terms of the sites you target and for how you use the data. If you display jobs from the optional remote boards, credit the source board, keep the attribution text, and link back to the job URL.

FAQ

Which dataset rows are free? Every job row costs $0.002. Other rows are free (no job event is charged): the one {"type": "status", "status": "warning", ...} row written when the demo, or a run stopped by its time limit, ends without any job row, the status: "unresolved" rows for company names or URLs that matched no board, and the status: "expired" rows that includeExpired adds. Job rows have no type field and no status field (with includeExpired on, they get status: "active"), so keep rows without status, or with status: "active", if you only want jobs.

What happens if the sample board in the demo is down or slow? The run still ends SUCCEEDED. Each request has a 10 s timeout in the demo (40 s otherwise) and only a couple of retries, and no new request is started about 40 s into the demo or 20 s before the run's timeout. If nothing could be read, the dataset gets one free record such as {"type": "status", "status": "warning", "message": "...", "source": "..."} with the reason, and the details are in RUN_SUMMARY. This record is not charged; filter on type if you only want data rows. A run on your own boards where every board fails still ends FAILED, so a wrong slug is easy to spot.

Which companies can I scrape? Any company whose public job board runs on Greenhouse, Lever, Ashby, Workable, Recruitee, Workday or Teamtailor. You don't need the company to be in the directory; the directory is only a shortcut for searching without a list. You can also opt in to the public Remote OK and Jobicy feeds with remoteBoards.

I only know the company name. What do I paste? Paste the name (Stripe, Hugging Face). The Actor checks its directory and a few slug spellings of the name on six ATSs (not Workday). If that finds nothing, you get a free unresolved record that lists the slugs tried. Then paste the company's careers page URL, or open it in a browser, follow the "view jobs" link and paste that board URL. Workday needs the full career site URL.

Does it cover Workday? Yes, when you paste a myworkdayjobs.com or myworkdaysite.com career site URL, or a careers page that links to or embeds one. Workday is not auto-detected from a bare slug and is not in the directory.

Why is SmartRecruiters not supported? Its API's robots.txt disallows automated access, and this Actor obeys robots.txt.

Do I need a login, cookies or an API key from the ATS? No. The Actor reads the public job board endpoints that the career pages themselves use. You only need an Apify account to run it. It obeys robots.txt and rate-limits requests per host.

How do I get only new jobs every day? Turn on onlyNewJobs and schedule the Actor. The first run returns everything that matches and remembers it. Later runs return only jobs that were not returned before, and you are charged only for those. Each run refreshes the jobs it still finds, so a job that stays listed is never returned twice; a job is forgotten after 120 days in which no run saw it. The state is kept in the named key-value store set by stateStoreName. Use a different store name for each independent scheduled feed. If the state store cannot be opened (for example an invalid stateStoreName, another account's store or missing permissions), the run still finishes but logs a warning, sets stateStoreUnavailable: true in RUN_SUMMARY and saves nothing. onlyNewJobs then cannot remember earlier jobs, so every matching job is returned and charged again on each run until the store works.

Why does a job have no salary? Many companies don't publish pay. When the ATS has a pay field the Actor uses it (salarySource: "ats"). Otherwise it looks for a range in the description text (salarySource: "description"), which needs includeDescription on, except on Workday, where the description is always read. Check salaryPeriod before comparing numbers, because some ranges are hourly or monthly.

What happens if a company slug is wrong or a board is closed? That entry is skipped with a warning in the log, and the rest of the run continues. You are not charged for it. If every entry fails, the run fails with an error message that lists the failed entries, so a bad input never looks like an empty but successful run.

Does the output contain personal data? The Actor returns job postings only. It has no recruiter or hiring-manager fields, and by default it removes personal e-mail addresses, phone numbers and "Your recruiter: name" lines from descriptions (redactContacts). Rule-based removal can miss unusual formats, so check before you republish descriptions.

Can I export to CSV or Google Sheets? Yes. Every run's dataset can be downloaded as CSV, Excel or JSON, and the Apify integrations for n8n, Make and Zapier can push rows to Google Sheets, Airtable, Slack and others.

Is this an API from the ATS vendor? No. It is an independent Actor that reads public job postings. It is not affiliated with or endorsed by the ATS vendors, or by Remote OK or Jobicy.

How should I credit jobs from the remote boards? Keep the attribution text on each remote-board item and link back to that item's job URL when you display the job. The boards ask for source credit; Remote OK also asks that the link be a follow link. Do not use a board's logo without permission.