Hiring Tracker: track job postings at companies (Lever, Ashby, Workable, Recruitee, Personio)

Track job postings at the companies you care about and see what is new or gone since your last run. Give the Actor a list of company domains or job board slugs and it returns one normalised record per open job, read from the public career site jobs on Lever, Ashby, Workable, Recruitee and Personio boards. Every job is marked new, existing or removed, so a scheduled run turns job postings into hiring signals and job change alerts. It suits sales and recruiting teams, analysts, investors and anyone running a competitor hiring monitor. Pricing is pay per result: $1.50 per 1,000 job records, and starting a run is free.

What's new

Hiring Tracker now reads Workable, Recruitee and Personio job boards as well as Lever and Ashby, so one Actor covers five ATSs. If you pass domains or board slugs, a company may now resolve to extra boards that were not found before. The jobs on those newly found boards are billed once as new on the first run after the upgrade, then show as existing on later runs as usual.

Use cases

• Sales prospecting from hiring signals. Watch target accounts and reach out when they open roles that point to a need, such as a first security hire or a new sales team.
• Competitor hiring monitor. Track which roles, teams and locations competitors are adding or closing, with the full job text if you want it.
• Investor and market research. Follow hiring at portfolio companies or a sector watchlist, and see where headcount is growing.
• Job boards and alerts. Feed a niche job board, newsletter or Slack channel with only the postings that appeared or disappeared since the last run.

Use it from Claude, Cursor or any AI agent

Add Hiring Tracker as a tool through Apify's MCP server: connect your MCP client to https://mcp.apify.com?tools=frostpine456/hiring-tracker with your Apify API token. Then ask things like "What roles is Ramp hiring for?" or "Which of these companies posted new jobs since last week?" For quick answers, pass one or two companies and turn includeDescriptions off.

Supported job boards

You can pass a company domain (acme.com) for any of these, or point at the board directly for a sure match. A bare slug (acme) is checked on all five systems.

Lever jobs scraper

Lever boards, including EU-hosted ones. Use https://jobs.lever.co/acme, https://jobs.eu.lever.co/acme or lever:acme.

Ashby jobs

Ashby boards. Use https://jobs.ashbyhq.com/acme or ashby:acme.

Workable jobs scraper

Workable boards. Use https://apply.workable.com/acme/, https://acme.workable.com or workable:acme.

Recruitee jobs

Recruitee boards. Use https://acme.recruitee.com or recruitee:acme.

Personio jobs export

Personio career pages, common among DACH companies. Use https://acme.jobs.personio.de, https://acme.jobs.personio.com or personio:acme.

Greenhouse, Workday, SmartRecruiters and Teamtailor are not supported.

Input

Each item in companies can be:

• a company domain: acme.com or https://www.acme.com/careers
• a board URL, including a link to one job on that board: https://jobs.lever.co/acme, https://jobs.eu.lever.co/acme, https://jobs.ashbyhq.com/acme, https://apply.workable.com/acme/, https://acme.workable.com, https://acme.recruitee.com, https://acme.jobs.personio.de (or .com)
• an explicit board: lever:acme, ashby:acme, workable:acme, recruitee:acme or personio:acme
• a bare slug: acme (probed against Lever US, Lever EU, Ashby, Workable, Recruitee and Personio; every system with a board for it is tracked)

Options:

• outputMode — all (default) outputs every open job each run; changesOnly outputs only new and removed records.
• includeDescriptions — set to false to omit descriptionText for smaller output.
• stateKey (advanced) — separates change-tracking state when you run several independent schedules.
• failOnEmpty (advanced) — health-check mode: the run fails if any company is unresolved, errors, or returns zero jobs.

Output

One dataset item per job:

{
  "company": "https://jobs.lever.co/palantir",
  "ats": "lever",
  "boardSlug": "palantir",
  "jobId": "a1b2c3d4",
  "title": "Software Engineer",
  "department": "Engineering",
  "team": "Infrastructure",
  "location": "New York, NY",
  "workplaceType": "hybrid",
  "employmentType": "Full-time",
  "compensationText": "USD 150000–200000 per year",
  "descriptionText": "We build ...",
  "postedAt": "2026-09-30T12:00:00.000Z",
  "url": "https://jobs.lever.co/palantir/a1b2c3d4",
  "applyUrl": "https://jobs.lever.co/palantir/a1b2c3d4/apply",
  "changeType": "new",
  "firstSeenAt": "2026-10-11T06:00:00.000Z",
  "lastSeenAt": "2026-10-11T06:00:00.000Z",
  "scrapedAt": "2026-10-11T06:00:00.000Z"
}

The Actor also writes a SUMMARY record to the key-value store with one entry per input company: status (ok / unresolved / error / skipped, where skipped means your charge limit was reached before that company was checked), boards found, job counts, and duplicates.

Pricing

You pay per job record (pay-per-event, one job event per dataset item): $1.50 per 1,000 jobs, with no charge for starting a run. For example, 1,000 job records cost $1.50. Warning: the first run bills every open job; use changesOnly on scheduled runs so later runs only bill jobs that appeared or disappeared since the previous run. Runs that find nothing for a company, and the SUMMARY record, are not billed.

FAQ

Is it legal, and what data does it collect?

The Actor reads public job postings only: no login, no personal data. Each record has the job title, department, location, employment type, pay text where the employer published it, the description, dates and links back to the original posting. Recruiter fields are not included, and email addresses and phone numbers found in free text are replaced with [redacted]. Whether a particular use is right for you is your call; this is not legal advice. Employers can ask for exclusion (see Takedown below).

How does changeType work?

The Actor remembers the jobs it saw on each board. On the next run with the same stateKey, a job that is new is marked new, one still open is existing, and one that has disappeared is removed. The first run has nothing to compare against, so every open job is new. Removed records carry no description.

How do I schedule it daily?

Save your input as an Apify Task, then open Schedules in the Apify Console and add a daily cron expression for that task. Keep the same stateKey between runs so changes are tracked. Use a different stateKey for each independent company list.

How do I get alerts only on new jobs?

Set outputMode to changesOnly. After the first run, each run returns only new and removed records, so you are billed only for those. The Actor has no built-in alerts: run it on an Apify Schedule and connect email, Slack or a webhook to the run.

What if a company is not found?

Its entry in the SUMMARY record is marked unresolved, and you are not billed for it. Domain resolution is best effort, so try the board URL (https://jobs.lever.co/acme) or the explicit form (ashby:acme). The company may also use a system that is not supported.

Does it support Greenhouse or Workday?

No. Supported sources are Lever, Ashby, Workable, Recruitee and Personio. SmartRecruiters and Teamtailor are not supported either.

Scheduling tips and limits

• Domain resolution is best effort. If you know the board, use its URL (https://jobs.lever.co/acme or ashby:acme) for certainty; bare domains may not resolve.
• Personio boards are read from the .de host only. A .com board URL is fetched from its .de twin, so a company that exists only on jobs.personio.com is reported as not found.
• Resolving domains takes about 1 to 3 seconds each because of polite rate limits (at most one request per second per host), so large domain lists take minutes.
• Removed jobs carry no description — they are reconstructed from the stored snapshot, which omits descriptionText to keep state small.
• State is kept per board, so adding a company to your list does not reset tracking for the others.
• A board that errors mid-run keeps its previous state, so the next run replays its changes instead of losing them.

Data and privacy

• Public job postings only — no login, no personal data.
• Email addresses and phone numbers found in free-text fields are replaced with [redacted].
• Every record links to the original posting via url (and applyUrl where available).

Disclaimer

Not affiliated with or endorsed by Lever, Ashby, Workable, Recruitee or Personio. Job content belongs to the posting employers.

Takedown

Employers can request exclusion via the Issues tab; we honour all requests.