open-jobs-data

A free, daily-updated dataset of open job postings from ~90 well-known tech companies, pulled directly from nine applicant tracking systems (ATS) and normalized into one schema.

Current snapshot: 15,919 postings across 90 companies (last full run: 2026-08-03, ~58s). Numbers move daily as postings open and close — see data/summary.json for the current count.

Data lives in data/:

• data/jobs.json — every posting, as JSON
• data/jobs.csv — same data as CSV
• data/summary.json — counts by platform/company, run stats
• data/README.md — schema doc

A GitHub Actions workflow (.github/workflows/update.yml) re-runs the fetcher every day and commits whatever changed. No manual updates.

Why this exists

Most companies with a real hiring pipeline publish their open roles through a public, unauthenticated JSON (or XML) feed on their ATS — Greenhouse, Lever, Ashby, and so on. These feeds are what each platform's own embeddable careers widget uses, so they're stable and meant to be read programmatically. This repo just fetches ~90 of them daily, normalizes the output, and commits it, so you don't have to write nine different parsers to get one CSV of open roles.

Schema

Field	Type	Notes
company	string	Display name
platform	string	greenhouse, lever, ashby, workday, smartrecruiters, workable, recruitee, personio, or bamboohr
jobId	string	Platform-native posting ID
title	string	Job title, as published
department	string | null	Not exposed by every platform
locations	array of strings	All listed locations
isRemote	boolean | null	Platform's own remote flag when present, else inferred from location text; null if neither is available
employmentType	string | null	e.g. "Full-time" — not exposed by every platform
applyUrl	string	Direct link to the application page
postedAt	string | null	ISO timestamp, when the platform reports one
scrapedAt	string	ISO timestamp of the fetch that produced this row

Sample row (from the live run above)

{
  "company": "Ramp",
  "platform": "ashby",
  "jobId": "34413f8d-26bf-4bbc-8ade-eb309a0e2245",
  "title": "Security Engineer, Cloud",
  "department": "Engineering",
  "locations": ["New York, NY (HQ)", "Remote (Canada)", "Remote (US)", "Miami, FL"],
  "isRemote": true,
  "employmentType": "FullTime",
  "applyUrl": "https://jobs.ashbyhq.com/ramp/34413f8d-26bf-4bbc-8ade-eb309a0e2245/application",
  "postedAt": "2026-04-07T17:12:35.753+00:00",
  "scrapedAt": "2026-08-03T19:18:46.567Z"
}

Using the data

curl

curl -s https://raw.githubusercontent.com/ConorsCode/open-jobs-data/main/data/jobs.json | jq '.[0]'

pandas

import pandas as pd
df = pd.read_json("https://raw.githubusercontent.com/ConorsCode/open-jobs-data/main/data/jobs.json")
df[df["isRemote"] == True].groupby("company").size().sort_values(ascending=False)

JavaScript

const jobs = await fetch(
  "https://raw.githubusercontent.com/ConorsCode/open-jobs-data/main/data/jobs.json"
).then((r) => r.json());

const remote = jobs.filter((j) => j.isRemote);
console.log(`${remote.length} remote postings`);

Coverage

90 companies as of the current snapshot, spanning all nine platforms below. Actual counts shift as feeds change — data/summary.json has the live breakdown.

Platform	Endpoint used	Companies in this run
Greenhouse	boards-api.greenhouse.io/v1/boards/{slug}/jobs	44
Lever	api.lever.co/v0/postings/{slug}	4
Ashby	api.ashbyhq.com/posting-api/job-board/{slug}	20
SmartRecruiters	api.smartrecruiters.com/v1/companies/{slug}/postings	5
Workable	apply.workable.com/api/v1/widget/accounts/{slug}	3
Recruitee	{slug}.recruitee.com/api/offers/	4
Personio	{slug}.jobs.personio.de/xml	2
BambooHR	{slug}.bamboohr.com/careers/list	2
Workday	{tenant}.{shard}.myworkdayjobs.com/wday/cxs/{tenant}/{site}/jobs	6

The full list, including the exact slug/board details used, is in companies.json.

Proposing an addition

Open a PR that adds an entry to companies.json. It needs a name, a platform (one of the nine above), and either a slug (for the bare-slug platforms) or a workday object with tenant, site, and shard (Workday board URLs don't follow a guessable pattern, so paste the tenant/site/shard from the company's actual board URL, e.g. https://intel.wd1.myworkdayjobs.com/External → tenant intel, site External, shard wd1).

Before opening the PR, confirm the endpoint actually returns postings — run node scripts/fetch.mjs locally after adding the entry and check data/summary.json's perCompanyStatus for your addition. Entries that come back empty or not-found will be reverted.

Companies are added on a rolling basis; there's no fixed cap, but each addition should be a real, well-known company with an active public feed, not a personal test account or an aggregator.

Limitations (read before relying on this for anything important)

• Only public, unauthenticated feeds. No login-gated boards, no scraping of rendered HTML, no LinkedIn-only postings, no custom in-house career pages that don't run one of these nine ATSes.
• No job descriptions. This dataset only pulls title/location/metadata fields, not the full posting body, to keep the daily fetch fast, cheap, and low-load on each platform.
• Snapshot timing. Each row reflects the moment scrapedAt records — postings can close or change within minutes of that. Don't treat a 27-hour-old row as still open without checking applyUrl.
• postedAt is always null for Workday and BambooHR. Neither platform's list endpoint exposes a real timestamp for it.
• isRemote is best-effort. It uses each platform's own remote flag when present, otherwise infers from location text containing "remote" — this can misclassify unusual location strings.
• An empty platform result for a company doesn't always mean "gone." Some platforms (Lever, Ashby, SmartRecruiters) return an empty list both when a company has zero current openings and, if the slug ever changed, when it's genuinely not found. data/summary.json records the raw per-company status so you can tell the two apart over time.
• No personal data. This dataset only contains what each ATS publishes about the job, never applicant data.
• Fixed company list, refreshed daily. This repo does not accept an arbitrary company list at request time — see below if you need that.

If you need this for your own company list

This repo's fetcher (scripts/fetch.mjs) is deliberately narrow: it's a compact, dependency-free script that only runs against the fixed list in companies.json, once a day, via GitHub Actions. That's what keeps it free and easy to audit.

If you want the same nine-platform normalization run against your own list of companies — with auto-detection of which ATS a bare company name uses, dedup across platforms when a company is findable on more than one, a per-company status report, and hosted execution so you don't run your own cron — that's a separate, paid tool: the ATS Jobs Scraper on Apify. It's pay-per-result (one result = one job posting) and covers the same nine platforms as this repo, plus optional full job descriptions.

This repo and that actor share no code — the actor is a proper TypeScript project with input validation, retries tuned per platform, and support; this repo is ~500 lines of plain Node meant to be read in one sitting.

License

MIT — see LICENSE. The code is MIT licensed; the job posting data itself is republished from each company's own public feed and belongs to the respective employer.