Jobs Feed API — Workday, Greenhouse, Lever & Ashby Jobs

Search live openings from thousands of companies' official Greenhouse, Lever, Ashby, Workable and Recruitee job boards. Deduplicated JSON with salary, remote and location filters. $3 per 1,000 jobs.

Title	Company	Locations	Workplace	Salary min
Software Engineer	Whitespace	Alexandria, Virginia, United States	remote	—
Staff Software Engineer, GI	Hinge Health	San Francisco-HQ	hybrid	197000
Senior Software Engineer (Fundraising) — Remote	GiveDirectly	Remote	remote	—

Real output from the “Remote software engineer jobs (last 7 days)” example, run on September 27, 2026.

Try a ready-made example: Remote software engineer jobs (last 7 days) · AI & machine learning engineer jobs · Product manager jobs in the United States · Data analyst jobs in Europe

Search live job openings from 10,000+ companies in one feed. The jobs come from the official public job-board APIs of Greenhouse, Lever, Ashby, Workable and Recruitee. Filter by keyword, location, country, remote, company, department, employment type and posting date. Keyword searches are ranked by relevance (title matches first), and you get flat, deduplicated JSON with salary ranges wherever the employer publishes them. $3 per 1,000 jobs, and you pay only for jobs delivered.

Quick start

1. Click Start with the prefilled input: keywords: ["engineer"], remote: "remote_only", postedWithinDays: 7, maxResults: 20. The run takes well under a minute.
2. You get the 20 most relevant remote engineering jobs posted in the last week (jobs with "engineer" in the title first, newest first among equals), each with company, locations, country codes, apply link and (where published) salary.
3. That first run costs at most $0.06 (20 jobs × $0.003), which fits in Apify's free monthly credit. Then change the filters or schedule the run with sinceLastRun: true to get a daily feed of new jobs only.

What you get

One flat record per job opening:

{
  "job_id": "ashby:ramp:34413f8d-26bf-4bbc-8ade-eb309a0e2245",
  "title": "Security Engineer, Cloud",
  "company_name": "Ramp",
  "company_board": "ramp",
  "ats": "ashby",
  "department": "Engineering",
  "team": "Backend",
  "employment_type": "full_time",
  "seniority": null,
  "workplace_type": "hybrid",
  "locations": ["New York, NY (HQ)", "Remote (Canada)", "Remote (US)", "Miami, FL"],
  "country_codes": ["CA", "US"],
  "remote": true,
  "salary_min": 211400,
  "salary_max": 290600,
  "salary_currency": "USD",
  "salary_period": "year",
  "salary_source": "structured",
  "years_experience_min": 5,
  "visa_sponsorship": null,
  "posted_at": "2026-04-07T17:12:35.753Z",
  "updated_at": null,
  "apply_url": "https://jobs.ashbyhq.com/ramp/34413f8d-26bf-4bbc-8ade-eb309a0e2245/application",
  "job_url": "https://jobs.ashbyhq.com/ramp/34413f8d-26bf-4bbc-8ade-eb309a0e2245",
  "description_text": "ABOUT RAMP\n\nRamp is building the smart infrastructure for finance teams…",
  "description_snippet": "ABOUT RAMP Ramp is building the smart infrastructure for finance teams, embedded in the transaction flow of every dollar a business spends…",
  "description_status": "included",
  "duplicate_sources": [],
  "match_score": 100,
  "matched_in": "title",
  "source_url": "https://api.ashbyhq.com/posting-api/job-board/ramp?includeCompensation=true",
  "fetched_at": "2026-09-19T03:00:00.000Z"
}

• workplace_type is one of remote, hybrid, onsite or unknown. We never guess: when an ATS has no workplace field (Greenhouse), the value comes from the location text ("Remote, Canada"), and otherwise it is unknown.
• Salary fields come from the ATS's structured compensation (Ashby, Lever, Recruitee, Personio) or, when it has none, from pay stated in the description text (salary_source says which). They are never estimated.
• country_codes are ISO 3166 alpha-2 codes. They come from the ATS's own country fields where available, and are otherwise inferred from the location text on a best-effort basis.
• duplicate_sources lists the other job_ids of the same opening. Example: a company that still has an old board live after switching ATS. Each opening is delivered and charged once.
• description_text is fetched live from the company's ATS for each delivered job (see below). description_status is included, not_requested (when includeDescription: false) or unavailable (the ATS could not be reached, or the posting has no text).
• match_score (0–100) and matched_in (title, department or description) show why a job matched your keywords. Both are null when you search without keywords.
• No personal data. Recruiter and hiring-manager fields are never copied. In descriptions, e-mail addresses, phone numbers and names in contact lines ("Questions? Call Jane on …") are replaced with [… redacted].

Enrichment (deterministic, no AI). Every job gets seniority (intern, entry, mid, senior, lead, manager, director, executive), read from the title by fixed rules, including common German, French, Spanish, Dutch and Portuguese markers. It is null when the title has no level marker ("Software Engineer"): no guessing. When the run reads the job's full description (includeDescription, the default), the text also gives years_experience_min ("3+ years", "3–5 years" → 3), visa_sponsorship (true when offered, false when ruled out, null when not mentioned) and, for jobs without structured pay, the pay stated in the text (salary_source: "description"; the ATS's own pay fields are structured and never overwritten). Without the description these three stay null. Personio publishes a level and the years of experience as fields: they fill seniority when the title has no level word and years_experience_min even without the description. Same input, same output: no language model, no estimates.

Use cases

• Job boards and aggregators: a clean, deduplicated feed of direct-from-employer jobs with apply links, refreshed daily.
• Sales and lead generation: see who is hiring for which team and where. A hiring burst signals budget. Filter by department ("sales", "security") or by company list.
• Recruiting and talent intelligence: track competitors' openings, salary ranges and remote policy.
• Job alerts and newsletters: schedule with sinceLastRun to get only the jobs posted since the previous run.
• AI agents and RAG: small input, flat deterministic JSON, and full plain-text descriptions (up to 20,000 characters).

Input

Field	Type	Notes
keywords	string[]	Case-insensitive. A title matches at a word start ("engineer" matches "Engineering Manager"). In the description, every word of the keyword must appear
keywordMatch	any | all	Default any
keywordScope	title | title_and_description	Default title_and_description: title, department, team and the first ~1,500 characters of the description
excludeKeywords	string[]	Drops jobs whose title contains any of them
locations	string[]	City/state/country/region text as written in the posting, or an ISO country code (US, DE, UK)
remote	any | remote_only | onsite_only	onsite_only excludes jobs marked remote
postedWithinDays	integer	First published within the last N days
companies	string[]	Company names (Datadog) or board tokens (datadog, greenhouse:datadog)
ats	string[]	Any of greenhouse, lever, ashby, workable, recruitee, workday, personio, teamtailor
departments	string[]	Substring match on department or team
employmentTypes	string[]	full_time, part_time, contract, temporary, internship, other
seniority	string[]	intern, entry, mid, senior, lead, manager, director, executive, read from the job title. Jobs whose title has no level marker are excluded when this is set
maxResults	integer	Default 100. With keywords, the most relevant jobs come first; without keywords, the newest
maxPerCompany	integer	Optional cap on jobs from any one company (prefill 3), so a single employer posting many near-identical roles can't fill your results
includeDescription	boolean	Default true. Fetches the full plain-text description (max 20,000 chars) live from the ATS for every delivered job. A 300-char snippet is always included. Set false for the fastest runs
sinceLastRun	boolean	Incremental feed: returns only jobs not delivered by an earlier run with the same filters
companyUrls	string[]	Live mode: board URLs (https://job-boards.greenhouse.io/gitlab, https://jobs.lever.co/acme, https://jobs.ashbyhq.com/acme, https://apply.workable.com/acme, https://acme.recruitee.com, https://acme.wd5.myworkdayjobs.com/External, https://acme.jobs.personio.de, https://acme.teamtailor.com; a Teamtailor custom domain as teamtailor:careers.acme.com) or ats:token strings. These boards are fetched live, so you can use companies that are not in the index yet

Ranking. With keywords, every matching job gets a match_score:

Where the keywords matched	match_score
Every keyword in the title	100
Some keywords in the title	60–70
Department or team	40–50
Description only	20–30

More keywords matched means a higher score inside each band. Among equal scores, the newest posting comes first. So a search for ["security engineer"] returns jobs titled "Security Engineer" before jobs that only mention security engineering in their description, however new those are. The Actor ranks across the whole index, not just the newest postings, before it delivers the top maxResults. Without keywords, results come newest first.

How it works. By default the Actor searches a compact, prebuilt search index of every board in our directory, rebuilt daily from the ATS APIs. The whole index is about 53 MB compressed. It holds every searchable field, plus the distinct words of each posting's first ~1,500 description characters. The Actor downloads only the parts that can match your ats, companies and date filters.

Descriptions. For every job it is about to deliver, the Actor fetches the description from the ATS. A single API call per company returns all of that company's jobs, so 100 jobs from 20 companies cost 20 calls. Lever and Workable calls are spaced out to respect their limits. A side benefit: a job the employer closed after the nightly build is detected and skipped, not delivered or charged. If a company's ATS can't be reached, the job is still delivered with description_status: "unavailable".

Live mode. With companyUrls it fetches those boards live instead of using the index.

Fallback. If the index is ever unreachable, the run falls back to live-fetching a small set of large boards, so a scheduled run never fails empty.

Incremental mode. With sinceLastRun: true, the Actor keeps a small cursor per filter combination in a named key-value store in your Apify account (factpipe-ats-jobs-feed-state). Jobs already delivered are skipped and not charged. A second run straight after the first returns 0 jobs and costs $0. Note that the cursor follows each job's posting date, so a board added to the index later only contributes jobs posted after your last run.

Pricing (pay per event)

Event	Price	Meaning
job-result	$3.00 per 1,000 ($0.003 each)	One job record delivered, with or without its description. No start fee. Runs with no matches, jobs skipped by sinceLastRun, jobs found closed at delivery time, and failed boards are never charged.

Example: 10,000 jobs cost $30. Set a maximum charge per run in Apify, and the Actor stops cleanly when it is reached.

Price change on November 6, 2026: job-result becomes $5.00 per 1,000 ($4.50 / $4.00 / $3.50 on Starter / Scale / Business plans), reflecting the added Workday, Personio and Teamtailor coverage and the new enrichment fields. Platform usage stays included.

Coverage

Search-index build of 2026-10-10:

ATS	Jobs in index
Workday	385,425
Greenhouse	148,443
Workable	79,741
Ashby	54,458
Lever	51,038
Recruitee	12,819
Total	731,924 from 15,339 company job boards (after collapsing 53,189 duplicate listings)

Workday sites contribute their newest 500 jobs each to the index; read a Workday site live (companyUrls) for all of its jobs.

Personio and Teamtailor (added 2026-10-14): 2,397 Personio and 1,956 Teamtailor career sites with open jobs (22,853 and 39,070 listed jobs) are in the directory and join the search index with its nightly build. Until a platform's first build, a search limited to it reads a few built-in boards live; any Personio or Teamtailor board also works in live mode (companyUrls).

About 15% of indexed jobs carry a structured salary range.

Coverage grows as the directory is refreshed. Every board in the index had at least one open job when it was last checked. If a company you need is missing, pass its board URL in companyUrls.

Sources and terms

Only official, public, unauthenticated job-board APIs and feeds that the ATS vendors publish for displaying and syndicating open jobs. No career-site HTML scraping, no logins, no LinkedIn or Indeed.

ATS	API	Reuse terms
Greenhouse	Job Board API	"Job Board data is publicly available"; no authentication for GET
Lever	Postings API	Published postings "are publicly viewable" and "may be scraped by third parties"; we respect its 1 request/s crawl delay
Ashby	Public Job Posting API	Public, unauthenticated posting API
Workable	Public jobs endpoint	Documented public endpoint for published jobs; robots.txt allows crawling
Recruitee	Careers Site API	Public API returning published offers
Workday	Public career sites (<company>.<wdN>.myworkdayjobs.com/<site>)	Not a vendor-documented API: the job list and job pages each site loads for its visitors. Each host's robots.txt is checked first (career sites allow /<site>/); about 4 requests/s across all Workday hosts
Personio	XML job feed (<company>.jobs.personio.de/xml)	Published by Personio for job portals and company sites. Each host's robots.txt is read first and only a text/plain file allowing /xml and /job/ counts; a company that disallows it is never read
Teamtailor	Career-site RSS feed (<company>.teamtailor.com/jobs.rss)	The jobs feed each career site links for its visitors (rel="alternate"), no Teamtailor API or login. robots.txt checked per host first (career sites disallow only /app/, /messages/ and similar)

SmartRecruiters is deliberately not included, because its API host's robots.txt disallows automated access. BambooHR is not included either: its Terms of Service forbid using any robot or spider to monitor or copy content from the service, which hosts its customers' careers pages.

factpipe Jobs & Hiring Data

These Actors search the same daily index of 730,000+ open jobs from 15,000+ company job boards:

• Jobs Feed API (this Actor): every job, every filter, ranked by relevance.
• Remote Jobs API: remote and work-from-home jobs only, filtered by region or time zone.
• Greenhouse, Lever, Ashby & Workday Jobs Scraper: pick the applicant tracking systems and companies you want.
• Companies Hiring: one row per company that is hiring for a role, as sales leads or market research.
• Greenhouse Jobs API: every Greenhouse job board in one search.
• Lever Jobs API: every Lever job board, with salary where published.
• Ashby Jobs API: every Ashby job board: startup and AI jobs, with salary where published.
• Workday Jobs API: thousands of Workday career sites of large employers (NVIDIA, Salesforce, Adobe…), with full descriptions.

FAQ

How is this different from LinkedIn or Indeed scrapers? We don't scrape any job site. Every job comes from the employer's own applicant tracking system, through the official public API that also powers the company's careers page. That makes the data first-party and current: a job is gone from the feed once the employer closes it. It also carries the direct apply link, and there are no reposts, recruiter spam or login walls. There is no ban risk and no proxy cost, which is why it costs $3 per 1,000 jobs.

How fresh is the data? The search index is rebuilt daily. With includeDescription on (the default), every delivered job is re-checked live against the employer's ATS during your run, and closed jobs are dropped. fetched_at is when the index read that board, and posted_at is the employer's own publish date.

Does keyword search cover the whole description? It covers the title, department, team and the first ~1,500 characters of each description, which is usually the role summary and the first requirements. Words that appear in most of one company's postings (its "About us" boilerplate) are matched per company. Use keywordScope: "title" for strict title searches.

Which companies are covered? About 15,000 company job boards with open jobs on Greenhouse, Lever, Ashby, Workable, Recruitee and Workday, plus the Personio and Teamtailor career sites added in October 2026 (see Coverage). They were discovered from Common Crawl's public URL index and each was validated against its ATS API or feed. Use companies to restrict the search, or companyUrls for any board that isn't indexed.

Why do some jobs have no salary? Only some employers publish pay: as structured ranges (mostly Ashby, Lever, Recruitee and Personio) or as a stated range in the description, such as US pay-transparency ranges on Greenhouse. The first is salary_source: "structured", the second "description" (read with fixed rules, only when the run reads the description). Pay is never estimated: a job that states none has null salary fields.

How are seniority, years of experience and visa sponsorship determined? By fixed, published rules, not by an AI model, so the same posting always gives the same answer. seniority comes from title words ("Senior", "Staff Engineer", "Head of", "Werkstudent"…); "Product Manager" or "Account Manager" are not people-manager titles and stay null unless the title has another level word. years_experience_min takes the first required "N+ years" in the description and ignores company-age phrases ("founded 20 years ago"). visa_sponsorship is false whenever the posting rules sponsorship out, even if it also mentions it positively. Anything ambiguous stays null.

Can I get only new jobs every day? Yes. Schedule the Actor with sinceLastRun: true. Each run returns and charges only jobs that no earlier run with the same filters delivered.

Can I call it from Python, JavaScript, Make, Zapier or an AI agent? Yes. Run it through the Apify API or the official Python/JavaScript clients. You can connect it to Make, Zapier, n8n, Slack or Google Sheets through Apify integrations, or expose it to AI agents through the Apify MCP server.

Related factpipe Actors

• Career Page Scraper: give it a company homepage and it finds the careers page and ATS for you.

Reliability

Official ATS APIs only. The Actor uses retries with backoff, per-host rate limits and structured failure reporting, and writes a RUN_SUMMARY record on every run. Its daily auto-test input has a built-in live fallback.