Jobs Feed: Every New Job from 45k+ Company Career Sites, Daily

A bulk feed of new job postings for data pipelines. Schedule it once a day and each run delivers every job that was added to our index of about 2 million jobs from 49,000 company career sites since your previous run: each job once, with no gaps, in one clean format.

Why this one

• Exactly once. The feed keeps a position, like a bookmark. A run that stops at its job limit (or your cost limit) is picked up by the next run at the exact job where it stopped, so nothing is skipped and nothing arrives twice.
• Built for volume. 100,000 jobs in a single run in about a minute, because runs read a ready-made daily index instead of crawling websites.
• Primary sources. Jobs come from the employers' own applicant tracking systems (19 of them, from Greenhouse, Workday, Lever and Ashby to BambooHR, Paylocity, Personio and Teamtailor), with working apply links, not reposts from job boards.
• Simple filters that keep a feed focused: countries, title keywords, remote/hybrid/on-site and ATS.

How the feed works

1. First run: delivers the matching jobs whose posting date is in the last N days (default 1; 0 = every matching job in the index). Jobs on platforms that publish no posting date have no known age, so they are not part of the first run; the new ones arrive with later runs.
2. Every later run: delivers every matching job that was added to the index since the last job delivered before, in the order the jobs were added. The index is rebuilt once a day, so one run per day picks up each day's new jobs; schedule it in the morning (UTC). When the index starts reading a company's career site for the first time, that company's older openings are new to the index too. Later runs: skip jobs posted more than N days ago (default 14) leaves those out, so the feed carries fresh postings; the status message says how many were left out. Set it to 0 to receive them as well. Jobs without a posting date are always delivered.
3. Max jobs per run caps a run. Anything over the cap is queued, not lost: the status message says how many are waiting, and the next run starts with them.
4. The feed's position is stored per set of filters in a key-value store named jobs-feed-state in your account. Set a Feed name to keep the position when you change filters, or to run two feeds with the same filters.

On 31 October 2026 the index held 2,016,483 open jobs from 49,020 companies. A first run for jobs posted in the last day returned 17,062 jobs (10,850 in the United States).

Filters

Filter	What it does
Countries	United States, Germany, GB... from the ATS's country field or the location text (Austin, TX is US, München is DE)
Job title contains any of / none of	e.g. engineer, nurse, sales; exclude intern. Matches the start of words, case-insensitive.
Workplace type	remote, hybrid, on-site (on-site includes jobs with no stated type)
Applicant tracking systems	e.g. greenhouse, smartrecruiters; every run's SUMMARY lists the platforms in the current index

Other filters of our Job Search API (salary, company, department, location text, employment type) are accepted through the API with the same field names.

Example input

All new US and Canadian jobs every day, starting with the last 3 days:

{
  "countries": ["United States", "Canada"],
  "firstRunDays": 3,
  "maxJobs": 50000
}

Output

One row per job, the same columns in every row (empty values are null): title, company, ats, companySlug, jobId, location, locations, country, countryCode (ISO 3166), workplaceType, department, team, employmentType, postedAt, updatedAt, firstSeenAt (when the job was added to the index: the feed's order), lastSeenAt, salaryMin, salaryMax, salaryCurrency, salaryPeriod, salaryText, url, applyUrl, sourceUrl, scrapedAt. ats + companySlug + jobId is a stable key for upserts.

{
  "title": "Senior Software Engineer, Backend (AI Platform)",
  "company": "Affirm",
  "ats": "greenhouse",
  "companySlug": "affirm",
  "jobId": "8040386003",
  "location": "Remote US",
  "locations": ["Remote US"],
  "country": null,
  "countryCode": "US",
  "workplaceType": "remote",
  "department": "Infrastructure Platform Eng",
  "team": null,
  "employmentType": null,
  "postedAt": "2026-10-30T17:52:49Z",
  "updatedAt": "2026-10-30T21:15:12Z",
  "firstSeenAt": "2026-10-31T01:40:39Z",
  "lastSeenAt": "2026-10-31T01:53:41Z",
  "salaryMin": null,
  "salaryMax": null,
  "salaryCurrency": null,
  "salaryPeriod": null,
  "salaryText": null,
  "url": "https://job-boards.greenhouse.io/affirm/jobs/8040386003",
  "applyUrl": "https://job-boards.greenhouse.io/affirm/jobs/8040386003",
  "sourceUrl": "https://job-boards.greenhouse.io/affirm",
  "scrapedAt": "2026-10-31T06:15:51Z"
}

The run's SUMMARY record shows the feed's position, how many new jobs matched, how many were delivered and how many are queued, plus the index date and size.

Plugging it into a pipeline

• Schedule: Apify Schedules, once a day.
• Deliver: integrations or a webhook on "run succeeded" push each run's dataset to Google Sheets, BigQuery, S3, Airtable, Make, Zapier or n8n, or fetch it from the dataset API as JSON, CSV or Excel.
• Upsert: key rows on ats + companySlug + jobId.
• Need the full job ad? The feed carries the index fields only, which keeps big runs fast. For the jobs you need, our Job Search API Actor fetches descriptions live (filter by company or job board).

Use cases

• Job boards and aggregators: a daily stream of fresh, primary-source jobs for your country or niche.
• AI agents and RAG: a clean, deduplicated daily delta to embed instead of re-crawling.
• Labour-market and hiring-signal analytics: new openings per company, role, country and ATS, every day.
• Recruitment marketing and sales intelligence: which companies started hiring for which roles today.

Pricing

You pay per job delivered plus a small fee per run; the exact prices are on the Pricing tab. Jobs your filters leave out and jobs delivered by earlier runs are never charged. Max jobs per run and the run's maximum cost both cap a run, and the queue keeps the rest for the next run.

FAQ

What counts as a new job? A job that appeared in the index after the last job your feed delivered, unless its posting date is older than Later runs: skip jobs posted more than N days ago (14 days by default). A job seen again on the next day is not new. A job the employer reposts under a new id, or takes down and lists again later, arrives again.

What if I miss a few days? Nothing is lost. The next run delivers everything added since your last run (up to Max jobs; the rest follows in later runs).

Can I change filters? Yes. Without a feed name, new filters start a new feed (first-run rules apply). With a feed name, the position is kept and the new filters apply from there on.

Is this legal? The index is built only from the public job-board feeds that ATS vendors provide so companies can publish their openings. No logins, nothing bypassed, no candidate or personal data. The job ads belong to the employers. This Actor is not affiliated with any ATS vendor or with any employer in the results.