Free. Scrape job postings straight from a company's own careers page, in seconds. No login, no proxies, no headless browser.
Most job scrapers go through job boards, which means stale listings, missing roles and rate limits. This one talks directly to the applicant tracking system (ATS) behind the careers page — the original source, updated the moment a recruiter posts a job.
Paste a career page URL or a company name. It works out which ATS the company uses and returns the jobs.
pip install -r requirements.txt
apify run| ATS | Coverage | Example input |
|---|---|---|
| Greenhouse | Full, incl. descriptions | greenhouse:stripe or https://boards.greenhouse.io/stripe |
| Lever | Full, incl. descriptions | https://jobs.lever.co/spotify |
| Ashby | Titles, locations, type | https://jobs.ashbyhq.com/ramp |
| SmartRecruiters | Full, incl. departments | smartrecruiters:Visa |
| Workday | Full (needs the complete URL) | https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite |
These are the companies' own public job board APIs. No proxies, no browser, no blocking — so a run costs almost nothing, and the Actor is free to use.
Everything here is a plain HTTP request. No key, no login.
import requests
# Greenhouse — one GET, add ?content=true for full descriptions
requests.get("https://boards-api.greenhouse.io/v1/boards/stripe/jobs?content=true").json()["jobs"]
# Lever — descriptionPlain gives clean text, no HTML stripping
requests.get("https://api.lever.co/v0/postings/spotify?mode=json").json()
# SmartRecruiters — paginated, 100 per page
requests.get("https://api.smartrecruiters.com/v1/companies/Visa/postings?limit=100").json()["content"]
# Workday — POST, and limit CANNOT exceed 20 (higher returns an empty array, not an error)
requests.post(
"https://nvidia.wd5.myworkdayjobs.com/wday/cxs/nvidia/NVIDIAExternalCareerSite/jobs",
json={"appliedFacets": {}, "limit": 20, "offset": 0, "searchText": ""},
).json()["jobPostings"]Ashby uses an unauthenticated GraphQL POST — see src/main.py for the query.
limitabove 20 returns an empty array, not an error. It looks exactly like "no more results."- It throttles fast paging. If your loop treats a failed page as the end of the list, you'll silently collect a fraction of the jobs. This cost us 1,960 of NVIDIA's 2,000 jobs before we noticed. Pause between pages and retry a failed one.
- The URL parts aren't guessable. The tenant, datacenter (
wd5) and site name all come from the company's real careers URL.
- Job title, company, location
- Department and employment type, where the ATS provides them
- Posted date, normalized across all five platforms
- Seniority level and job function, inferred from the title
- Direct apply link
- Full description text (optional; Greenhouse and Lever)
{
"company": "Careem",
"ats": "greenhouse",
"title": "Account Manager I",
"location": "Dubai, United Arab Emirates",
"department": "Commercial",
"employmentType": null,
"postedDate": "2026-06-24",
"seniorityGuess": "Mid-Senior level",
"functionGuess": "Sales",
"url": "https://boards.greenhouse.io/careem/jobs/1234567"
}The default input is a curated list of sought-after employers hiring in the UAE and Saudi Arabia. Every slug was checked against the live ATS, so you get real jobs on the first run.
| Company | Where | Input |
|---|---|---|
| Aldar Properties | Abu Dhabi | lever:aldar |
| Careem | Dubai | greenhouse:careem |
| Tamara | Dubai, Riyadh | greenhouse:tamara |
| Masdar | Abu Dhabi | smartrecruiters:masdar |
| Deliveroo | Dubai | ashby:deliveroo |
| Fresha | Riyadh, Dubai | lever:fresha |
| Lean Technologies | Riyadh | ashby:leantech |
| Kitopi | Dubai | lever:kitopi |
Add "locationFilter": "Dubai" (or Riyadh, Abu Dhabi, Remote) to narrow any run.
Most big employers aren't reachable. Of 42 top Gulf companies we checked, only 7 run a modern ATS. Emirates Group, ADNOC, Emaar, Majid Al Futtaim, Qatar Airways, DP World, Emirates NBD and Aramco all use Taleo, Oracle Cloud, Phenom or custom portals, none of which publish a public job API. For those, use the GCC Jobs Aggregator instead, which reaches them via Bayt, GulfTalent and NaukriGulf.
| Field | Description |
|---|---|
companies |
Career page URLs, company slugs, or ats:slug to force a platform |
titleFilter |
Only keep jobs whose title contains this text |
locationFilter |
Only keep jobs whose location contains this text |
includeDescription |
Fetch the full description for each job |
maxJobsPerCompany |
Cap results per company. 0 means no limit |
Mix formats freely:
{
"companies": [
"greenhouse:stripe",
"https://jobs.lever.co/spotify",
"https://jobs.ashbyhq.com/ramp",
"smartrecruiters:Visa"
],
"locationFilter": "Remote"
}Every Apify Actor is exposed over MCP, so you can wire this into Claude Desktop, Claude Code or any MCP client and ask: "what engineering roles are open at Stripe and Ramp right now?" It also has a REST API and works with LangChain, n8n, Make and Zapier.
Two bugs that shipped before we caught them, both from naive substring matching on job titles:
"coo" in title.lower()classified "Coordinator" as a chief operating officer."partner" in title.lower()promoted "Partner Onboarding Specialist" to firm partner.
Also: "Assistant Vice President" is not an executive — in Gulf banking and real estate it's an upper-mid individual contributor title. Use word boundaries, and print the raw titles behind every bucket before you trust an aggregate.
Some sites also fingerprint your TLS handshake, not just the User-Agent. When a site 403s a plain requests call but loads fine in Chrome, curl_cffi with impersonate="chrome" fixes it. (None of the five ATS platforms above need this — but the Gulf job boards do.)
These endpoints are public because companies want their jobs discovered. Cache aggressively, pause between requests, and don't hammer a small company's board. The reason this works without keys is that nobody has had to lock it down yet.
- GCC Jobs Aggregator — Bayt, GulfTalent and NaukriGulf in one search, deduplicated
- UAE Hiring & Lead Signals — which UAE companies are hiring now, with hiring-intent scores
MIT — see LICENSE.