Greenhouse, Lever, Ashby & Workable Jobs API

Pull every public job posting from companies' Greenhouse, Lever, Ashby and Workable job boards into one normalized JSON feed, with one dataset item per job.

The Actor calls only the unauthenticated job-board JSON APIs those products already publish for careers sites. It does not log in, does not use proxies, and does not call paid job APIs.

Why use it

• One input list of boards, one output schema
• Board URLs are enough — the Actor detects the ATS from the host
• Bad slugs become error rows; the rest of the run continues
• Optional keyword, location, department, remote, and recency filters
• Pay-per-event: a small start fee plus a fee per job item

How to get a board slug

Open the company's public careers page and copy the URL.

ATS	Careers URL	Slug
Greenhouse	https://boards.greenhouse.io/gitlab or https://job-boards.greenhouse.io/gitlab	gitlab
Lever	https://jobs.lever.co/spotify (EU boards use jobs.eu.lever.co)	spotify
Ashby	https://jobs.ashbyhq.com/ramp	ramp
Workable	https://apply.workable.com/soundcloud	soundcloud

You can paste that URL, or pass {"ats": "greenhouse", "slug": "gitlab"}.

Input

Field	Type	Description
companies	array	Required. Board URLs and/or {ats, slug} objects.
keywords	string[]	Keep jobs that contain at least one keyword (case-insensitive substring). Searches the title only unless keywordsMatch says otherwise.
keywordsMatch	title | title_and_description	Where keywords are searched. Default title. See Keyword matching.
locations	string[]	Keep jobs whose location text contains at least one value.
remoteOnly	boolean	Keep only remote roles.
departments	string[]	Match department or team text.
postedSince	integer	Keep jobs posted or updated in the last N days. Jobs with no date are kept.
maxJobsPerCompany	integer	Cap matching jobs per board after filters. Default 5.
includeDescription	boolean	Write cleaned Markdown (or plain text) on descriptionMarkdown. Default false.
descriptionFormat	markdown | text	Used only when includeDescription is true.
maxConcurrency	integer	Boards fetched at once. Default 3.

The default input is three well-known public boards with maxJobsPerCompany: 5 and descriptions off, so a first run stays small.

Example

{
    "companies": [
        "https://job-boards.greenhouse.io/gitlab",
        "https://jobs.lever.co/spotify",
        { "ats": "ashby", "slug": "ramp" },
        { "ats": "workable", "slug": "soundcloud" }
    ],
    "keywords": ["engineer", "designer"],
    "locations": ["Remote", "New York"],
    "remoteOnly": false,
    "departments": ["Engineering"],
    "postedSince": 30,
    "maxJobsPerCompany": 25,
    "includeDescription": false
}

Keyword matching

keywordsMatch controls where keywords are searched:

• title (default): the job title only. engineer returns "Backend Engineer" and "Engineering Manager", but not an "Account Executive" whose description mentions "solutions engineers".
• title_and_description: the title plus the full job description. This is how keywords worked before 1.0.8. It finds more jobs but also more off-target ones, and it is slower because descriptions have to be downloaded.

Inputs from before 1.0.8 still validate and run. To get the old matching back, add "keywordsMatch": "title_and_description".

Output

Each successful item looks like this:

{
    "status": "ok",
    "id": "greenhouse:gitlab:1234567",
    "company": "GitLab",
    "ats": "greenhouse",
    "slug": "gitlab",
    "title": "Senior Backend Engineer",
    "department": "Engineering",
    "team": null,
    "location": "Remote",
    "remote": true,
    "employmentType": "Full-time",
    "compensation": {
        "min": 150000,
        "max": 200000,
        "currency": "USD",
        "interval": "year"
    },
    "postedAt": "2026-10-01T20:00:00+00:00",
    "updatedAt": "2026-10-14T15:55:28+00:00",
    "applyUrl": "https://boards.greenhouse.io/gitlab/jobs/1234567",
    "jobUrl": "https://boards.greenhouse.io/gitlab/jobs/1234567",
    "descriptionMarkdown": null,
    "sourceId": "1234567"
}

compensation is null when the board does not publish a range. descriptionMarkdown is filled only when includeDescription is enabled.

If a company cannot be read, the Actor writes an error row and continues:

{
    "status": "error",
    "reason": "Board not found (HTTP 404) at https://boards-api.greenhouse.io/v1/boards/not-a-real-board/jobs?content=true&pay_transparency=true.",
    "company": "not-a-real-board",
    "ats": "greenhouse",
    "slug": "not-a-real-board",
    "sourceId": null
}

The dataset Overview view shows status, title, company, ATS, location, remote, dates, and links.

Pricing

Pay-per-event. See the pricing table in the Actor listing for the live figures. The published events are:

Event	Price	Charged
run-start	$0.002	Once per run
job	$0.0015	Once per job item

Error rows are not billed as job events. If you set a max total charge on the run, the Actor stops when the next event would exceed it and keeps the items already written.

Public APIs used

ATS	Endpoint
Greenhouse	GET https://boards-api.greenhouse.io/v1/boards/{token}/jobs?content=true
Lever	GET https://api.lever.co/v0/postings/{company}?mode=json (EU: api.eu.lever.co)
Ashby	GET https://api.ashbyhq.com/posting-api/job-board/{name}?includeCompensation=true
Workable	GET https://apply.workable.com/api/v1/widget/accounts/{subdomain} (?details=true when descriptions are needed)

These endpoints are public, unauthenticated, and intended for careers-site widgets. Identify your traffic and do not poll a board more often than you need.

Reliability

• Retries with exponential backoff on HTTP 429, 5xx, timeouts, and network errors
• Invalid or unknown slugs become error rows; the run does not fail
• Modest default concurrency (3 boards at a time)
• Charge-limit stop is clean: no partial charges beyond the cap the platform allows

Limitations

• Only the four ATS hosts above. Workday, SmartRecruiters, Lever-hosted custom domains, and Greenhouse embed-only sites without a public board token are out of scope.
• A board must be public. Private Greenhouse boards, Lever sites that are not published, and Workable accounts that disabled the widget return 404 and an error row.
• Greenhouse has no official remote flag. Remote is inferred from location and office text.
• Compensation is included only when the ATS publishes a structured range (most common on Ashby; occasional on Lever and Greenhouse). Workable's widget feed does not include pay.
• Lever createdAt is used as postedAt. Greenhouse uses first_published / updated_at. Workable often has a date without a time.
• Descriptions are HTML from the ATS, converted to Markdown or plain text. Conversion is best-effort; unusual markup can look messy.
• Large boards are downloaded in one or a few HTTP responses. maxJobsPerCompany limits how many items you write, not how much JSON the ATS returns. Leave includeDescription off if you only need titles.
• This Actor does not deduplicate across runs.
• Results only go as far as each employer published on that board. The Actor does not search the open web.

FAQ

Do I need API keys?
No. These job-board endpoints are public.

Can I pass a Greenhouse job URL, not the board URL?
Yes. https://boards.greenhouse.io/gitlab/jobs/123 is treated as the gitlab board. The same applies to a single Lever, Ashby, or Workable posting URL.

What happens if one company is wrong?
That company becomes an error row. Other companies still run.

Why is compensation null?
Most employers do not publish a structured range on the public board API. The Actor does not guess salary from description text.

Why is a role not marked remote?
Only an ATS remote flag, a workplaceType of remote, or location text such as "Remote" / "Work from home" counts. "Hybrid" is not remote.

Will the default input stay fast?
The prefills are public boards with maxJobsPerCompany: 5 and descriptions off. Run time still depends on how large those boards are on the day you run.

How do I keep costs down?
Set maxJobsPerCompany, add filters, leave descriptions off, and set a max total charge on the run.

More actors from this developer

• PDF to Markdown: DOCX, PPTX, XLSX, OCR & RAG Chunks: PDF, DOCX, PPTX, XLSX and HTML to clean Markdown and RAG-ready chunks.
• Google Trends Scraper & API: Interest, Related, Trending: interest over time, by region, and related queries from Google Trends.
• Google News Scraper: Full-Text Articles to Markdown: Google News articles as Markdown and RAG chunks for any topic or company.
• Website Screenshot & URL to PDF API (Bulk, Full Page): full-page and mobile screenshots (PNG, JPEG, WebP) and PDFs of any URL.
• Wayback Machine API: Snapshots & Archived Pages: every archived capture of a URL or domain, the snapshot closest to a date, and archived pages as Markdown.

Local development

python -m venv .venv
source .venv/bin/activate
pip install -r requirements-dev.txt
pytest

On the Apify platform the Actor starts with python -m src. Publish pay-per-event events run-start ($0.002) and job ($0.0015) to match this code. There is no deploy step in this repository.