Greenhouse Jobs Scraper

Yes. The same run reads Greenhouse, Lever, Ashby and SmartRecruiters boards and returns them in one 27-field schema.

Yes, for Greenhouse, Lever, Ashby and SmartRecruiters boards. This Actor costs $1.50 per 1,000 jobs on the free plan and $1.05 on paid plans; fantastic-jobs/career-site-job-listing-api costs $12.00, and its feed version $2.50 and $2.00. It reads each employer's board live through the official ATS APIs. Fantastic Jobs covers 58 applicant tracking systems against 4 here, so it wins if you need Workday, iCIMS and the rest.

Two ways. For a fixed workflow, add Apify's n8n or Make node, choose **Run Actor**, enter aurelix/greenhouse-jobs-scraper and map the dataset to the next step. For an AI agent (Claude, ChatGPT, Cursor or an n8n AI Agent node), add https://mcp.apify.com/?tools=aurelix/greenhouse-jobs-scraper as an MCP server and ask for a company's open jobs in plain words; setup for each client is in the section above. Runs bill your own Apify account at the same per-result price.

$1.50 per 1,000 jobs ($1.05 on paid Apify plans), billed through your Apify account; you pay only for results. Apify's free plan includes $5 of credit a month, enough for about 3,333 jobs. A price change is scheduled; the Store page shows it before it applies.

No. Type the company's domain (stripe.com): the Actor reads its public careers page (obeying robots.txt) and uses the Greenhouse, Lever, Ashby or SmartRecruiters board it links to; failing that it tries the domain's name on each ATS and keeps the board only when it points back to the domain. A bare name like 'Stripe' takes the first ATS with a board of that name, so prefer the domain. The run log shows which board was used. If nothing is found, paste a job link from the careers page (e.g. jobs.lever.co/spotify) or ats:slug (greenhouse:stripe).

It reads only public data any logged-out visitor can see: the public job-board APIs of Greenhouse, Lever, Ashby and SmartRecruiters that employers use to show jobs on their own sites. To find a board from a domain it reads the company's public careers page and obeys robots.txt. It goes at a polite pace, never logs in and never solves CAPTCHAs. It returns job posting fields (title, location, pay, dates, description, apply link); personal emails in descriptions are removed and no candidate or recruiter profiles are collected. To have a record removed, open an issue on this Actor's Issues tab; if an employer or job-board provider formally asks us to stop, this Actor will stop. You are responsible for using the data lawfully, incl. anti-spam and privacy laws; this is not legal advice.

The Actor remembers job IDs it has returned for the same companies and filters (or for the state name you set) and returns only unseen ones. Jobs are forgotten 60 days after they were last seen.

From the ATS country field where present, otherwise from the location text: 'Canada', province names and codes (ON, QC, BC...) and major Canadian cities. 'San Francisco, CA' is California, not Canada.

Salary is returned only where the employer publishes a pay range (common for Canadian and US pay-transparency jurisdictions). We never guess.

Yes. The input fields are described for agents: give 'companies' and optional filters. It works through Apify's MCP server and API.

No. Those have no official public job-board API; this Actor uses only official endpoints. More sources may be added as separate sister Actors.

Each job written to the dataset is one job charge; nothing else is charged per job. A board with no jobs that pass your filters costs nothing beyond the small per-run start fee. If no entry in Companies resolves to a board, the run fails at once with 'no supported job board found'. If one board errors (for example a token that no longer exists), it is skipped with a warning in the log and the others carry on; only when every board fails is the run marked failed.

Every job has a job_id made of the ATS, the board token and the posting id, lower-cased. Inputs that point at the same board (a domain and its board link, or a board that is also in the starter list) are merged before scraping, and each job_id is returned once per run. To skip jobs returned by earlier runs, turn on 'Only new jobs since last run'.

There is no run-wide cap: every open job on each board is returned unless you set 'Max jobs per company'. Boards are read four at a time. In our test runs, 4 boards took about 4 s and 25 boards returned 691 jobs in 27 s. SmartRecruiters needs one extra request per job for its description, so large SmartRecruiters boards are slower; switch descriptions off for those. To cap spend, set the run's maximum charge in Apify and output stops there.

Yes, unless you name the feed. The memory is keyed on the boards plus Canada only, Remote only, Title keywords and Countries, so changing any of those starts a fresh memory and the next run returns everything again. 'Max jobs per company' and 'Include full description' are not part of the key. Set 'Incremental state name' to keep one feed while you change the companies or filters.

workplace_type is remote, hybrid or onsite, taken from the ATS's own field where it has one (Lever, Ashby and SmartRecruiters) and otherwise from the location or metadata text; it is empty when nothing says. remote is the ATS's own remote flag where it has one (Ashby and SmartRecruiters), otherwise true when the workplace type is remote. employment_type is normalised to FULL_TIME, PART_TIME, CONTRACT, TEMPORARY, INTERN or OTHER, and employment_type_raw keeps the employer's own wording.

Any country an employer posts in. countries holds ISO codes from the ATS country field or the location text, and stays empty when no country can be detected; province tags are only for Canada. Titles and descriptions come in the language the employer wrote them, and Title keywords is a plain text match, so use the words of the posting's language (for example 'ingénieur' for French postings).

It uses no proxy: it calls the public job-board APIs of Greenhouse, Lever, Ashby and SmartRecruiters directly, at most four boards at a time, and backs off on HTTP 429 and server errors. When you give a company domain, it reads up to five pages of that company's public site to find the board, and only pages robots.txt allows.

Yes. Put a Slack incoming-webhook URL in 'Slack webhook URL', or any URL in 'Webhook URL' for Zapier, Make or n8n. After each run it posts the number of jobs saved (new ones only, when new-only mode is on) and the top 10 with links. Nothing is sent when the run saved nothing.