Multi-ATS Job Board Scraper

Scrape and normalize public job postings from Greenhouse, Lever, Ashby, and Workday into one clean dataset.

Use it for hiring-intent lead generation, recruiting intelligence, competitive hiring research, sales prospecting, and job-market monitoring.

What this Actor does

Give the Actor a list of ATS job board URLs or company domains. It discovers supported career pages, extracts public job postings, deduplicates them, filters by keywords, and returns standardized job records.

Supported ATS sources:

• Greenhouse — https://boards.greenhouse.io/<company>
• Lever — https://jobs.lever.co/<company>
• Ashby — https://jobs.ashbyhq.com/<company>
• Workday — common myworkdayjobs.com and myworkdaysite.com public career pages
• Domain discovery — scans a company homepage, /careers, and /jobs for supported ATS links

Common use cases

Use case	Example
Hiring-intent lead generation	Find companies hiring for RevOps, Salesforce, procurement, AI engineer, or data engineer roles
Recruiting intelligence	Track companies hiring for specific functions, locations, or seniority
Competitor monitoring	Watch competitor job boards for new teams, geographies, and product areas
GTM research	Build account lists based on current hiring signals
Job market analysis	Normalize postings across several ATS platforms into one dataset

Example input

{
  "startUrls": [
    "https://boards.greenhouse.io/airbnb",
    "https://jobs.lever.co/spotify",
    "https://jobs.ashbyhq.com/ashby",
    "https://adobe.wd5.myworkdayjobs.com/external_experienced/jobs"
  ],
  "domains": [],
  "keywords": ["engineer", "sales", "procurement"],
  "keywordFilterMode": "basic",
  "maxItems": 100,
  "maxItemsPerSource": 25,
  "includeDescriptions": true,
  "continueOnSourceError": true
}

Input options

Field	Type	Description
startUrls	array	Direct public ATS board URLs. Best for predictable scraping.
domains	array	Optional company domains. The Actor scans homepage, /careers, and /jobs for supported ATS links.
keywords	array	Optional keywords for filtering jobs. Leave empty to return all jobs.
keywordFilterMode	string	title, basic, or full. Controls where keywords are matched.
maxItems	integer	Maximum total number of jobs returned.
maxItemsPerSource	integer	Optional per-board cap so one large company cannot consume the whole run.
includeDescriptions	boolean	Include job description text in output. Disable for smaller/faster datasets.
continueOnSourceError	boolean	Continue scraping other sources when one source fails.

Keyword filter modes

Mode	Fields searched	Best for
title	Job title only	High-precision lead lists
basic	Title, department, location	Balanced precision and recall
full	Title, department, location, description	Maximum recall

Output fields

Each dataset item contains:

{
  "company": "spotify",
  "source": "lever",
  "job_id": "08d82da1-0c72-44d1-91eb-7fa22dee3d39",
  "title": "Analytics Engineer II",
  "department": "Platform",
  "location": "Stockholm",
  "remote": false,
  "url": "https://jobs.lever.co/spotify/...",
  "posted_at": "2026-05-13T15:23:52.322000Z",
  "description_text": "...",
  "matched_keywords": ["engineer"]
}

The Actor also writes source-level health information to the key-value store record SOURCE_STATUS, including per-source success/failure and item counts.

Pricing

This Actor uses pay-per-event pricing. You are charged per normalized job item returned in the default dataset.

Tips for best results

• Use direct ATS URLs when you know them.
• Use maxItemsPerSource when scraping several large employers.
• Use keywordFilterMode: "title" for precise lead lists.
• Use keywordFilterMode: "full" when you want maximum recall across descriptions.
• Disable includeDescriptions if you only need lightweight lead data.

Notes and limitations

• The Actor only extracts publicly available job postings.
• Workday career URL formats vary; common public Workday job pages are supported and paginated.
• Domain discovery depends on supported ATS links being visible from the scanned pages.
• Greenhouse public job APIs may expose updated_at instead of the original posting date.