Career Site Jobs Search - 9 ATS, Only New Jobs, No Duplicates

Search jobs by keyword across the career pages of thousands of companies at once. Type what you are looking for (data engineer, remote) -> get clean job rows straight from each company's own hiring page, in about 1 to 3 minutes, and on a schedule only the jobs that are new since your last run.

It reads the public job-board feeds of nine hiring platforms: Greenhouse, Lever, Ashby, Workable, SmartRecruiters, Recruitee, Personio, Breezy HR and Workday. No login, no cookies, no LinkedIn or Indeed scraping.

What you get

Field	Meaning
title, company, companySlug	Job title and the company that posted it
location, city, region, country, countryCode	Where the job is
workplaceType, isRemote	remote, hybrid or onsite (read from the board, or from the title and location)
seniority	intern, junior, mid, senior, staff, lead, manager, director or executive, read from the title
employmentType, department, team	full_time, part_time, contract ... and the team
salaryMin, salaryMax, salaryCurrency, salaryPeriod	Parsed salary range when the company publishes one
postedAt, updatedAt	When the job was posted, when the board gives a date
url, applyUrl, boardUrl	The job page, the apply link, and the company's board
source	Which platform it came from (greenhouse, lever, ashby ...)
matchedKeyword	Which of your keywords matched
isNew	true when running in "only new" mode
description	Full cleaned text, only when you switch "Include job descriptions" on

3 ways to use it

1. Find jobs for a keyword {"keywords": ["data engineer"], "maxResults": 200}
2. Remote senior roles posted this week {"keywords": ["backend", "platform engineer"], "locations": ["remote"], "seniority": ["senior+"], "postedWithinDays": 7, "maxResults": 500}
3. A daily fresh-jobs feed (pay only for new jobs) {"keywords": ["product designer"], "onlyNew": true, "feedName": "designer-eu", "locations": ["berlin", "london"], "maxBoards": 3000, "maxRuntimeMinutes": 20, "maxResults": 0} and schedule it daily.

Input

Field	What to enter	Default
keywords	Job words. Any of them may match; a keyword with several words needs all its words in the title, department or team	software engineer
excludeKeywords	Drop jobs containing these (intern, sales ...)	none
matchMode	All words in any order, or exact phrase	all words
searchDescriptions	Also search the job text (slower, more matches)	off
locations, countries, remoteOnly	Where. Type remote in locations to include remote jobs	anywhere
seniority, employmentTypes	Level and contract type filters	all
postedWithinDays	Only jobs posted in the last N days (jobs without a date are kept)	30
minSalary	Yearly minimum, for jobs that publish pay	0 (off)
maxResults	Stop after N jobs, newest first. This is also your spending cap. 0 = no cap	25
maxPerCompany	Cap jobs from any one company, so one employer posting a role in 40 cities does not fill the list. 0 = no cap	0
onlyNew, feedName	Return only jobs not delivered before; one feed name per saved search	off, default
companies	Search only these companies (name, domain, careers link or greenhouse:stripe). Clear keywords to list every job of a company	all known boards
platforms	Limit to some hiring platforms	all nine
maxBoards, maxRuntimeMinutes	How wide and how long to search	300 boards, 3 minutes
includeDescription, dedupe, webhookUrl	Descriptions, duplicate removal, Slack/Discord/Zapier/Make/n8n webhook	off, on, none
refreshBoardList, refreshBoardListDays	Re-discover company boards from the public Common Crawl index	off, every 30 days

The prefilled input works as-is: it returns 25 software engineer jobs.

Output example (a real row from a run)

{
  "title": "Delivery Solutions Architect - Public Sector (SLED)",
  "company": "Databricks",
  "source": "greenhouse",
  "location": "USCA",
  "workplaceType": null,
  "isRemote": false,
  "seniority": "lead",
  "salaryMin": 180000,
  "salaryMax": 247500,
  "salaryCurrency": "USD",
  "salaryPeriod": "year",
  "postedAt": "2026-10-01T14:20:08.000Z",
  "url": "https://databricks.com/company/careers/open-positions/job?gh_jid=8790285002",
  "boardUrl": "https://boards.greenhouse.io/databricks",
  "matchedKeyword": "solutions architect",
  "isNew": true
}

Download as JSON, CSV or Excel from the Storage tab.

Pricing (pay per event)

Event	Price
Job found (one job row delivered)	$0.002

You pay only for jobs that land in your dataset. If nothing matches, you pay nothing (you get one free notice row that explains why).

• 100 jobs = $0.20
• 1,000 jobs = $2.00
• A daily "only new" feed of 5,000 new jobs a week = $10 per week

Other career-site job APIs on the Store charge about $0.012 per job; this one is $0.002 and adds only-new mode.

Use it with

• Before / after: feed company names into that_red_bird/company-enrichment-by-domain to get each employer's profile; run that_red_bird/hiring-signal-monitor to watch a company's hiring over time; send rows to that_red_bird/lead-list-deduplicator to merge lists.
• Make, Zapier, n8n: schedule the Actor, trigger on "run succeeded", send rows to Google Sheets, Slack, Airtable or your ATS. Or set webhookUrl and get the new jobs pushed to you.
• AI agents / MCP: call that_red_bird/career-site-jobs-search with {"keywords": ["ml engineer"], "locations": ["remote"], "maxResults": 20} and the agent gets structured job rows with salary and seniority.

FAQ

How many companies does it search? It starts from a verified list of 5,700+ company career boards shipped with the Actor, and grows its own saved list from the public Common Crawl web index every 30 days. The default run reads the 300 biggest boards for speed; raise maxBoards (and maxRuntimeMinutes) to sweep thousands.

Why do I get fewer results than I expected? The default run is capped at 25 jobs and 300 boards. Raise maxResults, maxBoards and maxRuntimeMinutes. The run summary (SUMMARY in the Key-value store) says how many boards were read and how many were skipped because of the time limit.

How does "only new" work? Each feed name remembers the jobs it already delivered (in a named key-value store in your account). The first run returns everything that matches; later runs return only jobs not seen before. Use one feed name per saved search.

Are duplicates removed? Yes: the same company, title and location appearing twice is collapsed. This also catches a company posting on two platforms.

Is it legal? It reads the public job-board feeds that companies publish for job seekers and aggregators. Check the terms of any job data you republish.

Why is the salary empty? Only some companies publish pay. We parse it from the board's pay fields, and, with descriptions on, from the text. Nothing is guessed.

Do you cover LinkedIn or Indeed? No. Those sites block scrapers and are not covered. This Actor reads companies' own career sites.

Can a company be missing? Yes. Only companies on the nine supported platforms with a known board are searched. Add a company yourself with the companies field (name, domain or careers link).

Limits

• No LinkedIn, Indeed, Glassdoor or aggregator data.
• Nine platforms only; companies with a self-built careers page are not covered.
• Posting dates depend on the platform; Workday only reveals the day (or "30+ days ago").
• The board list is discovered from public web crawls, so very new companies can take a few weeks to appear. Add them with companies right away.

Support

Open an issue on the Issues tab; we answer quickly. If this saved you time, a short review helps other people find it.