Greenhouse Jobs Scraper: open jobs from any company's Greenhouse board

Give it the name or address of a company's Greenhouse job board and get one row per open job: title, department, location, pay range when the employer publishes one, posting date and a link. Filter by keyword, department, location and age, add full descriptions, get one hiring summary per company, or schedule it to return only jobs that are new since the last run.

Many companies host their job list on Greenhouse (boards.greenhouse.io/<name>). Greenhouse documents a public Job Board API for exactly this list. The Actor reads it, one request per board, and returns uniform records. It needs no login, no browser and no proxy.

For Workday, Lever, Ashby and other vendors in the same run, see the multi-vendor ATS Jobs Scraper by Tinlark in the section "More than Greenhouse?" below.

Who it is for

Recruiters and sourcers who track competitors' openings, sales and RevOps teams who read hiring as a buying signal, job-board and newsletter builders, and analysts of hiring trends.

How to find a company's Greenhouse board name

The board name (Greenhouse calls it the board token) is the last part of the board address. It is usually the company's name in lowercase.

1. Open the company's careers page and click an open job.
2. If the address looks like https://boards.greenhouse.io/stripe/jobs/1234567 or https://job-boards.greenhouse.io/figma/jobs/1234567, the board name is stripe or figma.
3. Some companies show jobs on their own domain and hide Greenhouse: the address may be careers.example.com/positions/1234567?gh_jid=1234567. The gh_jid in the address is the sign that Greenhouse sits behind the page. View the page source and search for greenhouse.io: an embed such as boards.greenhouse.io/embed/job_board?for=example gives the board name example. Paste that address or the name.
4. Not sure? Try the name. A wrong name gives a free error row that says so.

What you can paste

You paste	Result
stripe	board stripe (a bare name is looked up on Greenhouse)
greenhouse:stripe	the same, explicit
https://boards.greenhouse.io/stripe	board stripe
https://job-boards.greenhouse.io/figma	board figma
https://boards.greenhouse.io/embed/job_board?for=cloudflare	board cloudflare
https://boards.greenhouse.io/stripe/jobs/1234567	the whole board stripe, not just that job
figma.com	the Actor looks for a Greenhouse link on the company's home and careers pages, where robots.txt allows (best effort)

Domain lookup is hit and miss: among seven well-known company domains in Tinlark's tests, it found the board for two (Figma and Anthropic). It cannot see boards that a page loads with JavaScript. When it fails, use the board name.

What Greenhouse gives, and what it does not

Field	What it holds
company	The name set on the board by the employer
jobId	Greenhouse's job id, the number in the job address
title	The job title as posted
department	The first department of the job. Greenhouse allows several; only the first is returned
locations	The location text of the job, or its offices when no location text is set
isRemote	True when the location or title says remote. Greenhouse has no remote flag
postedAt	When the job was first published (UTC)
updatedAt	When the posting was last changed
salaryMin, salaryMax, salaryCurrency, salaryRanges	The pay range, only when the employer publishes one on the posting. salaryRanges keeps all ranges, with their labels, when there are several
jobUrl, applyUrl	The job's address. For companies that show Greenhouse jobs on their own site, this is that page (with gh_jid)
descriptionText	Only with includeDescription: the posting text as plain text
seniority	A guess from the title: senior, manager, intern and so on

Not available from Greenhouse's public board: team (stays empty), employment type, a workplace-type field, applicant counts, recruiter names and anything behind an applicant login. salaryInterval is empty when the pay label does not name one.

Pay ranges are common on some boards and rare on others. In a cloud run over ten large technology boards (3,572 jobs), 1,804 jobs carried a pay range. Do not expect that share on other boards.

Input

Field	What it does
companies (required)	Board names, greenhouse:name, board URLs or company domains
mode	jobs (default), signals (one summary row per company) or new-only
keywords, excludeKeywords	Matched against title, department and team, not case sensitive
departments	Text the department must contain, for example Sales or Engineering
locations	Text that a location must contain, for example Berlin or United States
remoteOnly	Keep jobs whose location or title says remote
postedWithinDays	Keep jobs first published within N days
includeDescription	Add the plain-text description to each row (much larger output)
maxJobsPerCompany	Newest first, default 25
maxItems	Stop after this many billable rows, default 1,000
stateName	A name such as sales-watch. Required for new-only. Remembers which jobs were seen
emitExistingOnFirstRun, includeClosed	Options for runs with a stateName

Example: sales jobs from three boards, last 14 days

{
  "companies": ["stripe", "airbnb", "https://boards.greenhouse.io/cloudflare"],
  "departments": ["Sales"],
  "postedWithinDays": 14,
  "maxJobsPerCompany": 100
}

Example: pay ranges only

Pay ranges are part of the output, not a filter. Run with maxJobsPerCompany set high and keep the rows where salaryMin is not empty, for example with a spreadsheet filter on the exported CSV. You do not need includeDescription for this.

Example: a daily watch

{
  "companies": ["figma", "databricks", "anthropic"],
  "mode": "new-only",
  "stateName": "ai-watch",
  "keywords": ["engineer", "research"]
}

The first run records the open jobs and returns none. Later runs return only jobs that are new.

Output

Results go to the default dataset (JSON, CSV, Excel, XML or the API). The SUMMARY record of the run's key-value store holds counts, errors and the number of requests.

Job row (real output from a cloud run)

{
  "recordType": "job",
  "atsVendor": "greenhouse",
  "company": "Airbnb",
  "boardToken": "airbnb",
  "jobId": "8247303",
  "title": "Software Engineer, Passport & Commerce, Android",
  "department": "Software Engineering",
  "locations": ["Remote, USA"],
  "isRemote": true,
  "postedAt": "2026-10-01T21:35:44Z",
  "updatedAt": "2026-10-01T21:35:44Z",
  "salaryMin": 162000,
  "salaryMax": 186000,
  "salaryCurrency": "USD",
  "jobUrl": "https://careers.airbnb.com/positions/8247303?gh_jid=8247303",
  "sourceUrl": "https://boards-api.greenhouse.io/v1/boards/airbnb/jobs",
  "status": "open"
}

Hiring summary (signals mode, trimmed)

{
  "recordType": "signal",
  "company": "Cloudflare",
  "boardToken": "cloudflare",
  "openRoles": 405,
  "rolesByDepartment": { "Solution Engineering": 69, "Engineering": 56, "Field Sales": 53, "Mid Market": 31 },
  "seniorRoles": 293,
  "remoteRoles": 3,
  "newLast7d": 36,
  "newLast30d": 161,
  "hiringSpike": false
}

Limits and honest notes

• Public boards only. Boards whose employer switched the public API off answer 404 and give an error row.
• No cap per board. Greenhouse returns the whole board, with posting texts, in one request; the largest board in Tinlark's tests had 885 jobs, and the peak memory of a ten-board run was 137 MB.
• The department is the first one if a job has several, and team is empty.
• Remote is a text match. A job in "Remote, USA" is remote; a job whose location says only a city is not, even if the employer allows remote work.
• jobUrl may not be on greenhouse.io. Some employers rewrite job addresses to their own career site.
• Closed jobs are reported only against your own earlier run with the same stateName. A board that suddenly returns no jobs after returning five or more is treated as an outage.
• Pace. At most 4 requests a second to Greenhouse's API, one request per board, retries with backoff on HTTP 429, 5xx and network errors, a declared TinlarkBot User-Agent.

Pricing

Free during launch (until 31 October 2026). You pay only Apify's own platform usage for your runs.

Measured platform usage: a cloud run over ten large boards returned 3,572 jobs in 6 seconds with a peak of 137 MB of memory and cost $0.018 in platform usage, about $0.005 per 1,000 jobs. The three-board prefill returns 75 jobs in about 5 seconds. The default memory is 512 MB.

From 1 November 2026: pay per event, $1.50 per 1,000 job rows ($0.0015 each on the Apify Free and Bronze plans; Silver $1.30, Gold $1.10 per 1,000) in jobs and new-only modes, $5.00 per 1,000 company summary rows in signals mode and $3.00 per 1,000 boards found from a company domain (not charged when none is found, and not for names or URLs). Error rows, closed-job rows and the baseline run of a new-only watch are free. For example, 50 boards with 40 jobs each is 2,000 jobs, or $3.00. Set Maximum cost per run to cap spending once prices are on.

More than Greenhouse?

This Actor is the Greenhouse-only version of ATS Jobs Scraper by Tinlark, which reads Greenhouse, Workday, Lever, Ashby, Workable, Recruitee and Personio in one run and tries a bare name on all of them. Both run the same code and return the same fields. A Workday-only version is Workday Jobs Scraper.

Use it from code or an AI agent

Start the Actor through the Apify API, any Apify client library or Apify's MCP server, and read the dataset items. The input and output schemas are defined, so tools can describe the fields themselves.

curl -X POST "https://api.apify.com/v2/acts/tinlark~greenhouse-jobs-scraper/run-sync-get-dataset-items" \
  -H "Authorization: Bearer <YOUR_APIFY_TOKEN>" -H "Content-Type: application/json" \
  -d '{"companies": ["stripe"], "departments": ["Sales"], "maxJobsPerCompany": 50}'

Data source, terms and your responsibility

The data comes from the Greenhouse Job Board API (boards-api.greenhouse.io/v1/boards/<name>/jobs). Greenhouse documents it as public: job board data can be read with GET requests and no authentication, so that companies and third parties can list postings. The Actor identifies itself as TinlarkBot, keeps to a gentle request rate and does not log in or read applicant data. The postings are the employers' own advertisements and carry no personal data about applicants. They belong to the employers: check their terms and the laws that apply to you before you reuse or republish them. Each row carries jobUrl and sourceUrl so you can link back.

This Actor is not affiliated with Greenhouse Software, Inc. or with any employer named above.

FAQ

Where do I find a board name? See the first section. In boards.greenhouse.io/<name> or job-boards.greenhouse.io/<name>, the name is the first part of the path.

Can I get salaries? Only where the employer publishes a pay range on the posting. Those rows have salaryMin and salaryMax. Greenhouse has no salary field for the other jobs.

Why did a company return an error row? Typical reasons: the board name is spelled differently from the company, the company uses another vendor, or the employer turned its public board off. The row says which request failed.

Can I run it for hundreds of boards? Yes. Each board is one request and the Actor sends at most four per second, so 300 boards take a little over a minute. The limit is 5,000 entries per run.

Is it legal to use this data? Disclaimers and legality. The Actor reads only what employers publish on their own public job boards, through the API that Greenhouse provides for that purpose, at a slow rate. It does not log in or get around blocks. The postings belong to the employers. You are responsible for how you use them, including the laws on unsolicited contact and on republishing content that apply to you. Tinlark is not affiliated with Greenhouse or any employer, and this is not legal advice.

Related Tinlark Actors

• ATS Jobs Scraper: all seven vendors in one run.
• Workday Jobs Scraper: the same fields for Workday career sites.
• Remote Jobs API: remote jobs from four job boards.

Support

Something wrong or missing? Open an issue on this Actor's Issues tab with your input (the board name or URL) and what you expected.