Hiring Signals by Domain: Greenhouse, Lever & Ashby Jobs API

Hiring Signals by company domain: give it a list of company domains and get open-job counts, departments, locations and posting dates from each company's public Greenhouse, Lever or Ashby job board (ATS jobs API), plus plain-word signals like hiring-sales.

Which of these companies are hiring, and for what? No LinkedIn, no login, no scraping of job-search sites: it reads the JSON feeds the ATS vendors publish for exactly this purpose.

One domain, before and after

Real output for ramp.com (run on 4 October 2026):

	Before	After
You have	ramp.com	ATS Ashby, board ramp, found via the link on ramp.com's own careers page
Open jobs	?	158 across 15 departments: 64 sales, 44 engineering, 24 ops/finance/people, 10 marketing
Freshness	?	52 posted in the last 30 days; newest today, oldest March 2023
Signal	?	hiring-sales, hiring-engineering
Where	?	New York (HQ), London, Remote (US), San Francisco, Toronto

A nine-domain run on 6 October 2026: Ramp (160 jobs, Ashby), Notion (133, Ashby), Linear (31, Ashby), Figma (159, Greenhouse) and Vercel (83, Greenhouse) were found through their own sites. Stripe (724, Greenhouse) and Spotify (74, Lever) do not link their board from their site; their boards were found by name and verified because the postings link to stripe.com and spotify.com (foundVia: slug-verified). example.com came back found: false: the Greenhouse board called example belongs to a different company and does not point back to it. A made-up domain came back found: false.

What it does

• Finds the board two ways: first the link on the company's own site and careers page (it cannot be the wrong company), then by trying the company name against the three ATS APIs. A guessed board counts when it points back to your domain: a posting links to it, the postings name it, or the board carries the company's name. Those rows say foundVia: slug-verified and verifiedBy says how. Unverified guesses are left out by default, because a common name can belong to a different company: the row says found: false with the board in guessedBoardUrl, and is not charged. Set includeGuessedBoards to count them as well; they are then marked foundVia: slug-guess.
• Companies mode (default): one summary row per company. Jobs mode: one row per open job with title, department, location, posted date and URL.
• Signals are simple keyword rules on department and title, so you can predict them: hiring-sales|engineering|marketing|product when a function is at least 3 jobs and 15% of the board; rapid-growth when at least half the dated jobs are under 30 days old on a board of 10 or more; small-team under 3 jobs.
• Companies without a board are returned as found: false and not charged. Lookups that fail with a network error are left out of the dataset, listed in the log and not charged.

Input example

The prefilled run:

{
  "companies": ["stripe.com", "ramp.com", "spotify.com", "notion.so", "linear.app",
                "figma.com", "vercel.com", "example-no-such-company-xyz.com"],
  "outputMode": "companies"
}

Field
companies	Domains (best) or company names
inputDatasetId, companyField	Read companies from another dataset; default column is the first of domain, website, company, name
outputMode	companies (default) or jobs
maxJobsPerCompany	Cap on job rows per company in jobs mode (default 200)
checkCareersPage	Read the company's own site for a board link before guessing (default on)
includeGuessedBoards	Also accept guessed boards that could not be verified against the domain (default off; needed for plain company names)
concurrency	Companies in parallel (default 3, kept low on purpose: the boards are public and shared)

Output example

A real companies-mode row for linear.app (run locally with this Actor's code on 5 October 2026):

{
  "type": "company",
  "input": "linear.app",
  "domain": "linear.app",
  "found": true,
  "ats": "ashby",
  "boardToken": "linear",
  "foundVia": "careers-page",
  "openJobs": 30,
  "departmentCount": 3,
  "jobsByFunction": { "engineering": 12, "product": 7, "other": 3, "sales": 3, "marketing": 2, "ops-finance-people": 3 },
  "topLocations": ["North America", "Europe", "London"],
  "newestPostedAt": "2026-09-15T18:05:47.890Z",
  "oldestPostedAt": "2021-04-27T20:13:45.158Z",
  "postedLast30Days": 3,
  "signals": ["hiring-engineering", "hiring-product"],
  "boardUrl": "https://jobs.ashbyhq.com/linear",
  "note": null
}

And a jobs-mode row from the same board:

{
  "type": "job", "input": "linear.app", "ats": "ashby", "boardToken": "linear", "function": "engineering",
  "id": "d3bc1ced-3ce4-4086-a050-555055dbb1ff", "title": "Senior / Staff Fullstack Engineer",
  "department": "Product", "location": "Europe", "postedAt": "2021-04-27T20:13:45.158Z",
  "url": "https://jobs.ashbyhq.com/linear/d3bc1ced-3ce4-4086-a050-555055dbb1ff"
}

Field	Meaning
type	company or job
input / domain	What you gave, and the domain taken from it (null for a plain name)
found	true if a Greenhouse, Lever or Ashby board was found
ats / boardToken / boardUrl	Which ATS, the board's name there, and its public URL
foundVia	careers-page (link on the company site), slug-verified (name matched and the board points back to the domain) or slug-guess (name matched, not verified: only with includeGuessedBoards)
verifiedBy	Only with slug-verified: what tied the board to the domain, e.g. a posting links to stripe.com
openJobs	Jobs listed on the board today
departmentCount	Distinct department names on the board
jobsByFunction	Jobs per group: sales, engineering, marketing, product, ops-finance-people, other
topLocations	Up to five most common job locations
newestPostedAt / oldestPostedAt	Newest and oldest posting date
postedLast30Days	Jobs posted in the last 30 days (null if the board gives no dates)
signals	hiring-sales, hiring-engineering, hiring-marketing, hiring-product, rapid-growth, small-team
guessedBoardUrl	Only with includeGuessedBoards off: a board that matches the name but could not be tied to the domain
note	Why a company was not found, or a warning on a guessed match
Jobs mode: function, id, title, department, location, postedAt, url	One open job

In jobs mode, companies without a board still get one free type: company row with found: false. A run summary (found, not found, failed, per ATS) is stored under the SUMMARY key.

What it does not do

• Only Greenhouse, Lever and Ashby. A company on Workday, Workable, BambooHR, its own page or nothing is found: false. Coverage is strong in tech and startups and weak elsewhere.
• Not a count of every job a company has: only what its public board lists today. Some companies list only a share of roles.
• Greenhouse department names are the company's own and can be fine-grained (Stripe: 197). Lever and Ashby posting dates are creation or publish dates, so a re-opened role can look old or new.
• It does not find people or email addresses.

Price

Price per 1,000 by your Apify plan (Apify applies the right column automatically):

	Free	Starter	Scale	Business and up
Company with a job board found, per 1,000	$4.00	$3.50	$3.00	$2.50
Job row returned, per 1,000	$0.50	$0.45	$0.40	$0.35

No start fee. The figures below use the Free-plan price; on a paid plan the same run costs less.

Pay per result, platform usage included:

Event	Price
Company with a job board found (companies mode)	$0.004
Job row returned (jobs mode)	$0.0005

• 1,000 companies found ≈ $4. 1,000 job rows ≈ $0.50.
• The seven-company run above, in companies mode: 7 boards found = $0.028; the made-up domain was free.
• Ramp in jobs mode: 158 job rows = $0.079. Stripe's 716 jobs are capped at 200 by the default maxJobsPerCompany, so $0.10.
• Free: companies without a board, failed lookups, and the found: false rows in jobs mode.

Set a maximum cost per run on the Apify run page; the Actor trims the list to fit rather than do work it cannot be paid for.

Use it from code or an AI agent

curl -X POST "https://api.apify.com/v2/acts/lyfe_tools~hiring-signals-by-domain/run-sync-get-dataset-items?token=YOUR_APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"companies": ["ramp.com", "linear.app"]}'

It returns the rows as a JSON array. AI agents (Claude, Cursor and other MCP clients) can call the Actor through the Apify MCP server at https://mcp.apify.com: the agent sends companies (and optionally outputMode) and reads found, ats, openJobs, signals and boardUrl.

FAQ

How do I find out which companies are hiring from a list of domains? Put the domains in companies. Each company gets a row with found, openJobs and signals.

Which ATS does a company use? The ats field says greenhouse, lever or ashby when a board is found. Other systems are not detected and show found: false.

Can I get every open job, not just a count? Yes, set outputMode to jobs. Each job is its own row with title, department, location, posted date and URL, up to maxJobsPerCompany per company.

Does it work with company names instead of domains? Yes, with includeGuessedBoards on: a name has no domain to verify against, so it can only be matched by guessing the board name, and such rows say foundVia: slug-guess. Domains are more reliable because the Actor can follow the link on the company's own site.

Does it scrape LinkedIn or Indeed? No. It reads only the public job-board APIs of Greenhouse, Lever and Ashby, plus the company's own homepage and careers page.

Am I charged when no job board is found? No. Only found boards (companies mode) or returned job rows (jobs mode) are charged.

Support

Something wrong, or a board it should have found? tools@lyfeoffshore.com. Reviews help others find the Actor.