Career Page Jobs Scraper: Company Name or Website to Jobs

Type company names or paste company websites. For each one this Actor finds the careers page, detects which applicant tracking system (ATS) hosts the job board, and returns the open jobs in one common format. It reads plain HTML and public job feeds only: no browser, no login, no proxy. It is unofficial and not affiliated with any ATS vendor or employer.

What does Career Page Jobs Scraper do?

For every company website, and for every company name after it has been matched to a website, the Actor:

1. Fetches the company's robots.txt and the homepage.
2. Looks for links to a supported ATS in the page source (links, iframes, scripts). If there are none, it follows up to a few same-site pages that look like careers pages (for example /careers, /jobs, "Join us", "Karriere", "Empleo") and looks there. A careers link that redirects straight to an ATS counts too.
3. If the pages show no board, it guesses (see "Name guess" below): the company name is tried as a board name on Greenhouse, Lever and Ashby.
4. Reads the open jobs from the detected board with the matching vendor reader.

Supported ATS (14): Workday, Greenhouse, Lever, Ashby, Workable, Recruitee, Personio, Dayforce, Eightfold, Zoho Recruit, ClearCompany (apply.clearcompany.com, formerly hrmdirect.com), Cornerstone, Jobvite and Rippling. If you already know the career-site address, paste it into companyUrls instead of the company website (for example https://jobs.lever.co/spotify, https://jobs.jobvite.com/nutanix/jobs, https://ats.rippling.com/rippling/jobs or https://apply.clearcompany.com/pmic) and it is read directly.

Every dataset item is one job with these fields: vendor, sourceUrl, company, jobId, title, location, locations, department, employmentType, remote, postedDate, salary (only where the employer publishes it), description (plain text, optional), jobUrl, scrapedAt, plus companyInput (the name or URL you gave), companyUrl (the website that was scanned, null when a board was found by name only), resolvedDomain (the website domain the input was matched to), careersPageUrl (the page where the ATS link was found), atsVendor, atsBoardUrl and detectionMethod (how the board was found: page_link, redirect, direct_url, domain_guess or name_guess).

A second output, the key-value store record DETECTION, lists for every input: the text you gave, whether it was a name or a URL, the status (detected, not_detected, robots_disallowed, blocked, fetch_failed), the resolved domain, the vendor, the detection method, the board URLs, the careers page and the number of jobs returned. It is free. The run's status message summarises it, for example "4 of 4 companies detected, 20 jobs".

How to run it

1. Open the Actor and keep the prefilled input: two company names (Figma, Palantir) and two company websites (Notion, NVIDIA), 5 jobs each.
2. Click Start. The prefill finishes in well under two minutes.
3. Open the Output tab for the jobs (table, JSON, CSV, Excel) and the key-value store for DETECTION.
4. Replace the input with the employers you track: type their names in companyNames, paste websites in companyUrls, or use both. Set maxItemsPerCompany to the number of jobs you want per company, and optionally searchQuery to filter by job title. One of the two lists must not be empty.

How are company names turned into job boards?

Names are resolved by fixed rules, without a search engine or any third-party lookup:

1. From the name the Actor builds up to six candidate domains: the name with spaces removed (Best Buy gives bestbuy.com), the hyphenated form (best-buy.com), and forms with a leading "The" and legal endings such as Inc, LLC, Ltd, GmbH, plc, Corp, Corporation, Company, Co, S.A. or AG removed. .com is tried first; .io, .ai and .co are tried only if no .com candidate was accepted.
2. A candidate is accepted only if its homepage answers with a success status (after redirects) and the page's <title> or og:site_name names the company: the site name, or a whole segment of the title such as the part after "|", must equal the name once punctuation, spaces and capitals are ignored. A page that merely mentions the name ("name.co for sale") is rejected. A homepage that redirects to a different domain is accepted only if that domain's name still fits the company name.
3. On the accepted website the normal careers-page and ATS detection runs. Items found this way carry detectionMethod: "domain_guess".
4. If no website is accepted, or the website shows no supported ATS, the name is tried as a board name on Greenhouse, Lever and Ashby (name_guess).

Homepages that answer 403 or similar are skipped, never retried with tricks. Names are matched to the company that owns the domain with that name; they can be wrong, see the FAQ.

How much does it cost to scrape company careers pages?

The price is 0.0022 USD per job returned (one result event per dataset item). 1,000 jobs cost 2.20 USD; 5 jobs from each of 4 companies cost 0.044 USD. Companies where no ATS is detected, or that block automated access, return no items and are not charged. Resolving a name costs nothing extra. You can cap spending with the maximum total charge setting of the run: the Actor stops as soon as the limit is reached and makes no further requests.

Input example

{
  "companyUrls": [
    { "url": "https://www.notion.com" },
    { "url": "https://www.nvidia.com" }
  ],
  "companyNames": ["Figma", "Palantir"],
  "searchQuery": "",
  "maxItemsPerCompany": 5,
  "includeDescription": true,
  "maxPagesPerCompany": 4,
  "guessBoardsFromDomain": true
}

Output example

Four items from local runs on 2026-10-14 (v0.4). Descriptions are shortened here, the dataset holds the full text. The first three come from the prefill input above: a website (page_link) and two names (domain_guess). The last one is the name "Block" in a run with includeDescription off: no website was accepted, so the board was found by name only (name_guess, no companyUrl, no resolvedDomain).

[
  {
    "vendor": "ashby",
    "sourceUrl": "https://jobs.ashbyhq.com/notion",
    "company": "notion",
    "jobId": "1fc309c8-da20-4ff2-84c7-8b863ece2b0a",
    "title": "Software Engineer, Developer Platform",
    "location": "San Francisco, California",
    "locations": [
      "San Francisco, California",
      "New York, New York"
    ],
    "department": "Engineering",
    "employmentType": "Full time",
    "remote": true,
    "postedDate": "2026-08-24",
    "salary": null,
    "description": "WHO WE ARE  Notion is the collaborative AI workspace where teams and agents think together https://w ...",
    "jobUrl": "https://jobs.ashbyhq.com/notion/1fc309c8-da20-4ff2-84c7-8b863ece2b0a",
    "scrapedAt": "2026-10-14T12:40:32Z",
    "companyInput": "https://www.notion.com",
    "companyUrl": "https://www.notion.com",
    "resolvedDomain": "notion.com",
    "careersPageUrl": "https://www.notion.com/careers",
    "atsVendor": "ashby",
    "atsBoardUrl": "https://jobs.ashbyhq.com/notion",
    "detectionMethod": "page_link"
  },
  {
    "vendor": "greenhouse",
    "sourceUrl": "https://boards.greenhouse.io/figma",
    "company": "Figma",
    "jobId": "5426468004",
    "title": "Account Executive, Enterprise",
    "location": "San Francisco, CA • New York, NY • United States",
    "locations": [
      "San Francisco, CA • New York, NY • United States"
    ],
    "department": "Sales",
    "employmentType": null,
    "remote": null,
    "postedDate": "2025-01-28",
    "salary": "USD 165,000 - 190,000 (Annual Base Salary Range:)",
    "description": "Figma is growing our team of passionate creatives and builders on a mission to make design accessibl ...",
    "jobUrl": "https://boards.greenhouse.io/figma/jobs/5426468004?gh_jid=5426468004",
    "scrapedAt": "2026-10-14T12:40:50Z",
    "companyInput": "Figma",
    "companyUrl": "https://figma.com/",
    "resolvedDomain": "figma.com",
    "careersPageUrl": "https://www.figma.com/careers/",
    "atsVendor": "greenhouse",
    "atsBoardUrl": "https://boards.greenhouse.io/figma",
    "detectionMethod": "domain_guess"
  },
  {
    "vendor": "lever",
    "sourceUrl": "https://jobs.lever.co/palantir",
    "company": "palantir",
    "jobId": "6ed76ce8-4156-4b60-b120-403538bd66cd",
    "title": "Administrative Business Partner",
    "location": "Singapore, Singapore",
    "locations": [
      "Singapore, Singapore"
    ],
    "department": "Administrative",
    "employmentType": "Full-time",
    "remote": false,
    "postedDate": "2026-08-11",
    "salary": null,
    "description": "A World-Changing Company   Palantir builds the world’s leading software for data-driven decisions an ...",
    "jobUrl": "https://jobs.lever.co/palantir/6ed76ce8-4156-4b60-b120-403538bd66cd",
    "scrapedAt": "2026-10-14T12:40:57Z",
    "companyInput": "Palantir",
    "companyUrl": "https://palantir.com/",
    "resolvedDomain": "palantir.com",
    "careersPageUrl": "https://www.palantir.com/careers",
    "atsVendor": "lever",
    "atsBoardUrl": "https://jobs.lever.co/palantir",
    "detectionMethod": "domain_guess"
  },
  {
    "vendor": "greenhouse",
    "sourceUrl": "https://boards.greenhouse.io/block",
    "company": "Block",
    "jobId": "5317296008",
    "title": "Account Manager, SMB",
    "location": "Sydney, Australia",
    "locations": [
      "Sydney, Australia"
    ],
    "department": null,
    "employmentType": null,
    "remote": null,
    "postedDate": "2026-07-09",
    "salary": null,
    "jobUrl": "http://block.xyz/careers/jobs/5317296008?gh_jid=5317296008",
    "scrapedAt": "2026-10-14T12:34:57Z",
    "companyInput": "Block",
    "companyUrl": null,
    "resolvedDomain": null,
    "careersPageUrl": null,
    "atsVendor": "greenhouse",
    "atsBoardUrl": "https://boards.greenhouse.io/block",
    "detectionMethod": "name_guess"
  }
]

A DETECTION record entry for a name looks like this:

{
  "input": "Palantir",
  "inputType": "name",
  "inputUrl": "https://palantir.com/",
  "resolvedDomain": "palantir.com",
  "status": "detected",
  "reason": "homepage of www.palantir.com is titled \"Home | Palantir\"",
  "vendor": "lever",
  "detectionMethod": "domain_guess",
  "boardUrls": ["https://jobs.lever.co/palantir"],
  "careersPageUrl": "https://www.palantir.com/careers",
  "jobsPushed": 5
}

How the two newer readers work: Jobvite career sites are read from their public listing pages (50 jobs per page, followed page by page) and one public job page per returned job, so department, location and description come from the job page; posting date and employment type appear where the site publishes them. Rippling boards are read from the board's public JSON listing; the listing has one row per job and location, and the Actor merges the rows of a job into one item with all its locations. With includeDescription on, one more public request per job adds the description, employment type, posting date and published pay range.

Name guess: what it is and how to turn it off

Many company websites do not show their job board in the page source. When the page scan finds no board, or no website was found for a name, and guessBoardsFromDomain is on (the default), the Actor takes the name (for a website: the domain name, shopify.com becomes shopify; for a name: its compact and hyphenated forms) and asks three public job APIs whether a board with that name exists: Greenhouse, Lever and Ashby. A board counts only if it lists at least one job. Greenhouse wins when its board name equals the company name; otherwise Lever, then Ashby. That is 3 to 4 extra requests per tried name, all robots.txt-checked GET requests.

A guess can be wrong: a name_guess board can belong to a different organisation with the same name. Items found this way carry detectionMethod: "name_guess", and the DETECTION record gives the board name, the number of jobs and the first job title so you can check it. To use only boards that a company's own pages link to, set guessBoardsFromDomain to false; then name_guess never appears (names that do not resolve to a website return nothing). You can also filter the dataset on detectionMethod.

Limitations and the free plan

• Names that are short, common words (for example "Square" or "Block") can belong to several organisations. The Actor accepts a website only when the page names the company exactly, but the owner of that domain may still be a different organisation: check resolvedDomain and jobUrl. Companies whose website blocks automated access, or whose robots.txt disallows the homepage, cannot be resolved by name through the website; only the name guess is left for them.
• Detection works only when the link to the ATS is present in the raw HTML of the homepage or of a careers page. Sites that build their job list with JavaScript, host their own job database, or use an unsupported ATS are reported as not_detected. Measured on 2026-10-14 with v0.4 on 20 company names chosen beforehand (US employers, tech companies of mid size, European companies, several with ambiguous names), the matches were: 13 found and correct, 0 found and wrong, 7 not found (the DETECTION record shows why). A second list of 10 names gave 5 found and correct, 0 wrong. In these runs 12 of the 18 correct matches came from the name_guess step (often after the website was found but did not show its board) and 6 from the website's careers page (domain_guess). Expect many companies to be not_detected.
• Jobvite and Rippling were added in v0.3. Jobvite sites on the company's own domain (for example a careers host that points at Jobvite) are only found when the page links to jobs.jobvite.com/<company>; older hire.jobvite.com addresses are not read. A Rippling board with no open roles returns no items. The name guess does not cover these two vendors.
• Sites that answer 403, 429 or 503, or show a challenge page, are skipped with status blocked. Nothing is retried and no proxy or header trick is used.
• robots.txt is enforced for every host that is fetched, including candidate homepages, the careers page host and the ATS host. Disallowed pages are not requested. A Crawl-delay is honoured.
• The Actor waits at least one second between requests to the same host and runs one request at a time. A company costs a few seconds to a minute depending on the vendor and on whether descriptions are included; the 20-name run above took 5 minutes in total (about 15 seconds per name on average, one name after another), so large name lists need a longer run.
• A company can have up to 3 boards; the most referenced board is read first, and maxItemsPerCompany applies to the company as a whole.
• Salary is filled only where the employer publishes it. Some vendors use the title for searchQuery filtering rather than a native search.
• The prefill returns 20 jobs and costs 0.044 USD at the per-job price, so the monthly credit of a free Apify plan can pay for it; larger runs are charged per job as above.

FAQ

Should I use names or URLs? Names are quicker to type and fit tools that only know the company name. Websites are the precise option: use companyUrls when a name could be ambiguous or when you already know the site. You can give both lists in one run.

How do I know a name matched the right company? Look at detectionMethod, resolvedDomain and jobUrl in each item. domain_guess means the name was matched to a website whose title names the company and the board was read from that site's careers page. name_guess means only a board with the same name was found: check that the job links lead to the company you meant. page_link, redirect and direct_url come from a website or career-site URL that you gave yourself.

Does it collect personal data? No. It reads job-ad fields (title, location, department, published pay, description). E-mail addresses found in job descriptions are replaced by [email removed], and recruiter or contact fields are never read.

Does it log in or solve captchas? No. Only anonymous, public pages and public job feeds are used.

Why was my company not_detected? Its ATS may be unsupported, its website may block automated access, or its job list is rendered by JavaScript. Pass the career-site address directly in companyUrls if you know it, or raise maxPagesPerCompany. The name guess only covers Greenhouse, Lever and Ashby boards named like the company.

Where do I report a problem? Use the Issues tab of this Actor on Apify.