Career Site Jobs Monitor: Greenhouse, Lever, Ashby, Workday, Workable, Recruitee, Personio, Teamtailor & BambooHR

Watch the career sites of any list of companies and get only the jobs that are new or changed since your last run, in one clean JSON schema. The actor reads each ATS's public job-board endpoints (the same ones the company's own careers page uses). No proxies, no logins, no browser, so runs are fast and cheap.

Try it free: Apify's free plan gives $5/month of usage, no card needed. That's about 5,000 jobs a month at $1.00 per 1,000.

Supported job boards

ATS	Short form	Board URL example	Notes
Greenhouse	greenhouse:gitlab	https://job-boards.greenhouse.io/gitlab	Structured salary when published
Lever	lever:palantir	https://jobs.lever.co/palantir	Spaced 1 s apart (Lever's crawl delay)
Ashby	ashby:ramp	https://jobs.ashbyhq.com/ramp	Structured salary
Workday	workday:adobe.wd5/external_experienced	https://adobe.wd5.myworkdayjobs.com/external_experienced	Up to 2,000 jobs per search (Workday's limit); see Workday search text
Workable	workable:huggingface	https://apply.workable.com/huggingface/
Recruitee	recruitee:bunq	https://bunq.recruitee.com/	Structured salary when published
Personio	personio:ottonova	https://ottonova.jobs.personio.de/	.de and .com boards
Teamtailor	teamtailor:polestar	https://polestar.teamtailor.com/	*.teamtailor.com boards
BambooHR	bamboohr:scribd	https://scribd.bamboohr.com/careers	Posted date comes from a per-job call
SmartRecruiters	smartrecruiters:<companyId>	https://jobs.smartrecruiters.com/<companyId>	Opt-in only, see FAQ

Don't know the board? Paste the company's careers page (for example https://www.figma.com/careers/) or just its domain. The actor reads that page, checks robots.txt first, and finds the job board it links to. Pages that load jobs only with JavaScript may not reveal their board; the run says so in OUTPUT.

What you can use it for

• Job alerts: schedule a monitor and send only new postings to Slack, Telegram, email, a Google Sheet or your own webhook.
• Niche job boards: feed a board (remote AI jobs, climate jobs, fintech) from hundreds of company career sites.
• Hiring signals for sales and recruiting: a new "Head of RevOps" posting, a changed salary band, or a role that closed tells you what a company is doing.
• Salary research: pay ranges become min, max, currency and period.
• AI agents and automations: a small, predictable schema for n8n, Make, Zapier, LangChain or an MCP client.

Features

• Ten ATS platforms behind one input and one output schema; mix them in one run.
• Monitor mode:
  • New jobs only: each scheduled run returns postings it has not returned before.
  • New and changed: also returns jobs whose title, location, department, employment type, remote status or salary changed (changeType: "updated"), and jobs that closed and came back ("reopened"). Description wording edits are ignored, so you don't get noise.
  • List jobs that closed: jobs that disappeared since the last run go to the CLOSED_JOBS record and the webhook. They are not billed.
  • State is kept per board, so adding a company to a running watchlist does not flood you (with skip jobs already open on the first run).
• Filters: title keywords (optionally the description too), excluded keywords, locations, remote only, departments, employment type, posted within N days, minimum yearly salary.
• Webhook delivery: saved jobs are POSTed as JSON in batches, with an optional secret header. No call when nothing is new.
• History dataset: optionally append every saved job to a named dataset that grows across runs.
• Respects Max jobs saved and your run's maximum charge. A job cut off by either limit is delivered on the next monitor run, never lost.
• One failing board never stops the run. Each board's count or error is in the OUTPUT record.

Input example

{
  "sources": ["greenhouse:gitlab", "workday:adobe.wd5/external_experienced", "workable:huggingface", "https://www.figma.com/careers/"],
  "keywords": ["engineer", "AI"],
  "excludeKeywords": ["intern"],
  "remoteOnly": true,
  "postedWithinDays": 14,
  "includeDescription": "none",
  "monitorMode": "newAndChanged",
  "monitorName": "remote-ai-eng",
  "reportClosedJobs": true,
  "webhookUrl": "https://example.com/my-hook",
  "maxJobs": 200
}

Output example

Real item from a run on 28 Oct 2026 (monitor fields added in v2):

{
  "id": "greenhouse:gitlab:8638232002",
  "ats": "greenhouse",
  "board": "gitlab",
  "company": "GitLab",
  "jobId": "8638232002",
  "title": "AI Transformation Owner, CRO",
  "department": "Enablement",
  "location": "Remote, United States",
  "country": "US",
  "isRemote": true,
  "workplaceType": "remote",
  "postedAt": "2026-07-22T17:38:40.000Z",
  "updatedAt": "2026-10-15T19:24:46.000Z",
  "url": "https://job-boards.greenhouse.io/gitlab/jobs/8638232002",
  "salary": { "min": 139200, "max": 235200, "currency": "USD", "period": "year", "source": "ats" },
  "scrapedAt": "2026-10-28T03:33:33.859Z"
}

In monitor mode each item also has isNew, changeType (new, updated or reopened) and firstSeenAt. Fields an ATS leaves empty come back as null.

What each ATS fills in

From our live test on 31 Oct 2026 (up to 20 jobs per board):

Board tested	Posted date	Location	Country	Description	Salary
greenhouse:gitlab	20/20	20/20	19/20	20/20	8/20
lever:palantir	20/20	20/20	20/20	20/20	14/20
ashby:ramp	20/20	20/20	20/20	20/20	20/20
workday:adobe.wd5/external_experienced	20/20	20/20	20/20	20/20	11/20
workable:huggingface	6/6	6/6	6/6	6/6	0/6
recruitee:bunq	20/20	20/20	20/20	20/20	0/20
personio:ottonova	9/9	9/9	0/9	9/9	0/9
teamtailor:polestar	13/13	13/13	13/13	13/13	0/13
bamboohr:scribd	2/2	2/2	2/2	1/2	0/2

Salary appears only when the company publishes it. Personio lists the office city without a country, so country is often empty there.

Pricing

$1.00 per 1,000 jobs saved on every plan, plus Apify's default actor start event ($0.00005 per run). Closed-job records and webhook calls are free.

Examples: a daily monitor that finds 40 new or changed jobs costs about $1.20 a month. A one-off pull of 5,000 jobs costs $5.00. Set Max jobs saved or a maximum charge per run to cap any run.

How to find a company's board

Easiest: paste the careers page URL and let the actor find it. Or look at a job link on the careers page:

• job-boards.greenhouse.io/<board> or boards.greenhouse.io/<board>: Greenhouse
• jobs.lever.co/<board>: Lever
• jobs.ashbyhq.com/<board>: Ashby
• <tenant>.wd5.myworkdayjobs.com/<site>: Workday (paste the whole URL)
• apply.workable.com/<account>: Workable
• <board>.recruitee.com, <board>.jobs.personio.de, <board>.teamtailor.com, <board>.bamboohr.com

A wrong board name is listed as BOARD_NOT_FOUND in the OUTPUT record and the other boards still run.

Schedule a job monitor

1. Set Monitor mode to New jobs only or New and changed, and give the watchlist a name.
2. Turn on skip jobs already open on the first run to start quietly.
3. Add a Webhook URL, or connect an Apify integration (Slack, email, Google Sheets).
4. Create an Apify schedule, for example every 6 hours.

Use it from n8n, Make or Zapier

Apify's official n8n, Make and Zapier integrations can run this actor on a schedule and pass the new jobs to Telegram, Slack, email or a sheet. With Monitor on, each run returns only new or changed jobs, so the message holds only what is new.

FAQ

How fresh is the data? Every run reads the live job board, so results match the careers page at run time.

Does it use proxies or a browser? No. These are the endpoints the ATS serves to career sites, so plain HTTPS requests are enough.

Can I monitor hundreds of companies? Yes. Boards are read 4 at a time by default (Boards read in parallel), with per-host delays. A failing board is skipped and reported.

Do I pay for jobs I have already seen? No. In monitor mode only new (or, if you choose, changed) jobs are saved, and only saved jobs are billed.

I used version 1 with "only jobs not seen before". Do I need to change anything? No. That setting still works and your existing watchlist carries over without re-sending old jobs.

Why can Workday show "2,000" jobs? Workday's job search stops at 2,000 results. For very large Workday sites, set Workday search text (for example engineer) so the list is narrowed before filters run. A large Workday site takes about a minute per 2,000 jobs, because pages are fetched politely one after another.

What about SmartRecruiters? SmartRecruiters publishes its Posting API for each company's own career site, and its robots.txt asks automated clients to stay out. The actor reads a SmartRecruiters board only when you turn on SmartRecruiters boards you are authorised to read, for example your own company's.

Data and compliance

The actor reads public job postings through endpoints each ATS serves for job boards and career sites, and it checks robots.txt before reading a careers page you give it. It does not log in, bypass any access control or collect candidate data. Postings can include a recruiter's name in the description text; you are responsible for using the data lawfully, including GDPR, CCPA or India's DPDP Act where they apply.

Limits

• Salary appears only when the company publishes it.
• Workday: at most 2,000 jobs per search (see FAQ). BambooHR and Workday need one extra request per job for descriptions and exact dates, so they are slower when you ask for descriptions.
• Careers-page detection finds boards linked in the page HTML. It can miss boards loaded only by JavaScript; in that case paste the board URL.
• Teamtailor boards on a custom domain: paste the *.teamtailor.com address.

Support

Open an issue on the actor's Issues tab with your input and the run link. Issues are answered within one business day.

Built with the help of AI coding agents. Version 2.0 is covered by 33 unit tests, 11 live checks against real boards and a 15-check end-to-end run, all passing on 31 Oct 2026.