Greenhouse, Lever, Ashby and Workday jobs API: the free endpoints and their gotchas

Greenhouse, Lever and Ashby each publish a public, keyless JSON feed of a company's open jobs. You need the company's board token, which is usually the last part of its job-board URL. Workday works differently, and has its own section below.

The endpoints

Gotchas

• No server-side date filter on any of them. Lever filters location, commitment, team, department and level on the server, but not dates. Download the board and filter locally.
• Use the right date field. Greenhouse's updated_at changes whenever a post is edited. Use first_published, which the list endpoint currently returns, though the docs only show it on the single-job endpoint. Lever's equivalent is createdAt (epoch milliseconds); Ashby's is publishedAt.
• Greenhouse pay ranges are per job. Call /jobs/{id}?pay_transparency=true to get pay_input_ranges.
• Lever has two hosts. An EU company returns 404 on api.lever.co, so try api.eu.lever.co.
• Ashby has unlisted jobs. Skip jobs where isListed is false.
• A bad token is a 404. All three returned 404 for a made-up token in our test on 22 Oct 2026. That means "wrong token or wrong ATS", not "not hiring".

import requests

def greenhouse(token):
    r = requests.get(f"https://boards-api.greenhouse.io/v1/boards/{token}/jobs", timeout=20)
    r.raise_for_status()
    return [{"id": j["id"], "title": j["title"], "location": j["location"]["name"],
             "posted": j.get("first_published"), "url": j["absolute_url"]} for j in r.json()["jobs"]]

def lever(token, eu=False):
    host = "api.eu.lever.co" if eu else "api.lever.co"
    r = requests.get(f"https://{host}/v0/postings/{token}?mode=json", timeout=20)
    r.raise_for_status()
    return [{"id": p["id"], "title": p["text"], "location": p["categories"].get("location"),
             "posted": p.get("createdAt"), "url": p["hostedUrl"]} for p in r.json()]

def ashby(token):
    r = requests.get(f"https://api.ashbyhq.com/posting-api/job-board/{token}", timeout=20)
    r.raise_for_status()
    return [{"id": j["id"], "title": j["title"], "location": j["location"],
             "posted": j.get("publishedAt"), "url": j["jobUrl"]}
            for j in r.json()["jobs"] if j.get("isListed", True)]

for name, jobs in [("greenhouse:stripe", greenhouse("stripe")),
                   ("lever:spotify", lever("spotify")),
                   ("ashby:ramp", ashby("ramp"))]:
    print(name, len(jobs), jobs[0]["title"] if jobs else "-")

For daily "new jobs" alerts, diff job IDs against yesterday's run rather than trusting dates, because reposted jobs get fresh dates. To find tokens, see how to scrape Greenhouse jobs without a browser.

Workday jobs API

Workday has no public job-board feed like the three above. Each career site ({tenant}.wd{N}.myworkdayjobs.com/{site}) loads its jobs from its own search endpoint, which you can call too: POST /wday/cxs/{tenant}/{site}/jobs with a JSON body. It isn't documented as a public API, so expect changes, keep requests slow and respect the site's robots.txt.

What trips people up:

• 2,000 results per search, 20 per page. After that, later pages repeat the first page, and total is capped at 2,000 (and is 0 on pages after the first). For bigger boards, split the search by the facets the site offers (job family, location) and merge the results by job ID.
• Posted dates are approximate. The listing only gives relative text such as "Posted 3 Days Ago", and older jobs show "Posted 30+ Days Ago". Converting that gets you to within about a day, at best.
• Some rows are empty stubs. Skip postings with no title or link.

import time, requests

def workday(host, tenant, site, search=""):
    url = f"https://{host}/wday/cxs/{tenant}/{site}/jobs"
    jobs, offset, total = [], 0, None
    while True:
        r = requests.post(url, json={"appliedFacets": {}, "limit": 20, "offset": offset,
                                     "searchText": search}, timeout=20)
        r.raise_for_status()
        data = r.json()
        if total is None:
            total = data["total"]           # only reliable on the first page
        page = data["jobPostings"]
        if not page:
            break
        jobs += [{"title": j["title"], "location": j.get("locationsText"),
                  "posted": j.get("postedOn"), "path": j["externalPath"]}
                 for j in page if j.get("title") and j.get("externalPath")]   # skip non-public stubs
        offset += 20
        time.sleep(1)
        if offset >= min(total, 2000):
            break
    return total, jobs

total, jobs = workday("TENANT.wd5.myworkdayjobs.com", "TENANT", "SITE")
print(total, "on the board,", len(jobs), "read")
print(jobs[:3])

We built this

Disclosure: Your Evenings Ltd built and sells this. Greenhouse, Lever, Ashby & Workday Jobs API is a pay-per-use Actor on the Apify Store that does all of this for a list of company domains:

• It finds the board on Greenhouse, Lever (including EU), Ashby, Workable, SmartRecruiters, Recruitee or Workday, from a domain, careers URL, board URL or ats:token.
• Workday: give it a company name, domain or myworkdayjobs.com link. It splits large boards to read past the 2,000-result cap. Each row's postedAtBasis says approximate-from-listing-text when the date came from Workday's relative text. Try the Workday jobs export example.
• Title, location, remote, department and posted-date filters run before you're charged.
• companyTotalJobs on every row and a COMPANY-STATUS file account for every company. monitorMode returns only opened or closed jobs. Companies it can't find are free.

It doesn't cover other systems (for example iCIMS or SuccessFactors) yet.

Longer walkthrough: Greenhouse jobs API: get jobs posted in the last 24 hours (and why your scraper returns 0 jobs) on dev.to.

FAQ

Is there a free Greenhouse jobs API? Yes. boards-api.greenhouse.io/v1/boards/{token}/jobs is public and keyless for reading published jobs.

Do Lever and Ashby need an API key for job postings? No. Both posting APIs are public for reading.

Does Workday have a public jobs API? Not a documented one. Career sites use their own search endpoint, which returns at most 2,000 results per search and shows only approximate posted dates.

Product names and logos belong to their owners. Your Evenings Ltd is not affiliated with or endorsed by them.