ATS Job Postings Scraper — Greenhouse, Lever, Ashby · billed only for new postings

30 days, 100 companies. A scraper that bills per row: ~40,000 rows a month → $60 at $1.50/1,000 — and 95% of those are postings you already saw. Here: ~2,000 genuinely new postings → $4. The more often you run it, the cheaper it gets.

Fetch job postings from companies' public ATS job boards and get them back in one consistent shape, whichever system the company uses.

Built for a specific question — did this company just start hiring — which makes it more useful to sales and lead-gen teams tracking that signal than to recruiters reading the postings themselves.

What makes it different

You only pay for postings you haven't seen before. The actor remembers what it delivered in previous runs and skips it — so a daily run returns today's new openings, not the same list over and over.

Companies covered

363 companies today — 272 Greenhouse · 77 Ashby · 14 Lever. Press Start and it runs against all of them. You don't need to know how any company configures its careers page.

Narrow it down, or go beyond the catalog:

Input	What it does
catalog_tags	Keep only companies matching a tag — tech, fintech, ai, data, design
catalog_sources	Keep only one ATS — lever, greenhouse, ashby
catalog_limit	Cap how many companies get queried in a single run
targets	Query any company by its board identifier, whether or not it's in the catalog: [{ "source": "lever", "company": "palantir" }]
max_postings	Hard cap on how many postings a run delivers — and therefore what it costs. Defaults to 2500 (about $5.00)

Output

Every posting comes back in the same shape, no matter the source:

{
  "source": "greenhouse",
  "company": "stripe",
  "job_id": "4567",
  "title": "Backend Engineer",
  "location": "New York, NY",
  "url": "https://boards.greenhouse.io/stripe/jobs/4567",
  "posted_at": "2026-08-01T10:00:00-04:00",
  "department": "Engineering",
  "description_html": "<p>…</p>"
}

What a real run looks like

Measured, not estimated — a run against the full catalog on 22 Aug 2026:

Companies queried	298
Postings found	55,325
Postings delivered	7,905
Time	5 minutes

The gap between found and delivered is the whole point: the other 47,420 had already been delivered in earlier runs, so they cost nothing. On your first run everything is new — after that, you pay for the difference.

Sources

All three sources are public, unauthenticated endpoints that their own documentation or robots.txt permits. Systems whose terms prohibit automated extraction aren't supported — that rules out Workday and SmartRecruiters.

How it behaves

• Identifies itself on every request, with a real contact address. No browser spoofing.
• Stays under published rate limits and backs off on 429 instead of hammering.
• A failing source never takes down the others — if one board is down, the rest still return.

Running it on a schedule

The point of paying only for new postings is running often. Open the actor's Schedules tab and create one — daily, weekly, your call. Run the actor manually at least once first: Apify requires a prior run before it lets you schedule one.

To be notified when a run brings something new, add an email or Slack notification from the actor's Integrations tab. Runs that find nothing new still finish quietly, so a notification tied to new results only reaches you when there is something to see.

---

Built by KHASSINX LLC.