Documentation [email protected]Log in

Scrape job postings into a dataset you define

Job postings data powers labor analytics, sales triggers and market maps — but vendors sell it as billion-row licenses. Quantic AI scrapes the slice you define instead: roles, companies, locations and posted dates from career pages and public boards, discovered through the Google Jobs vertical first.

$1 free every month · No card required · Blocked pages are never billed

hiring pipeline · example run
# "data engineer postings in Berlin, last 30 days"
search("data engineer berlin", jobs vertical)
→ postings + hiring companies · $0.0005
map(company sites) · find /careers, /jobs
→ career sections located · $0.0005 each
scrape(postings, extract: title, salary, date)
→ rows: { "title": "Data Engineer",
    "company": "…", "posted": "2026-07-14" }
Jobs verticalGoogle Jobs results as structured JSON
Career pagesthe primary source, mapped automatically
Your slicerole × geo × date window, your columns
No licensecents per run, pay per success

How it works

From a hiring question to a postings table

1 · Ask the market

"Data engineer postings in Berlin", "who's hiring sales in Milan fintech", "remote React roles posted this month". The SERP API's Google Jobs vertical returns postings and the companies behind them as structured JSON.

2 · Go to the source

For each hiring company, a map call finds /careers and /jobs sections — the postings' primary home, where detail is richest and freshest. No board markup in the way.

3 · Extract the posting

Each posting page is scraped with your schema via the scraping API: title, location, posted date, salary where stated, requirements. Rows are deduplicated across boards and career pages.

4 · Track the signal

Rerun weekly as a batch job and diff: new roles are sales triggers, closed ones are churn signals, counts by company map your competitors' growth — your slice, on your schedule.

A job posting data API for pipelines

Everything above is callable from code: the jobs vertical, map and scrape as REST endpoints or MCP tools — the same AI web scraping service pipeline that runs from a prompt. Labor-analytics and sales-intelligence teams can feed their models without negotiating a catalog license.

Batch jobs keep a fixed company list refreshed — up to 1,000 URLs per job, webhooks on completion, unused items refunded — and every response carries its own cost, so the data product's margin is visible per row.

Scrape job postings from the open web only

Postings are public by intent — companies publish them to be found. This pipeline honors the boundary anyway: career pages, public ATS pages and open boards yes; logged-in platforms no. No LinkedIn credentials, no walled-garden workarounds.

That boundary is also why the data is cleaner: a posting on the employer's own site has no aggregator lag, no repost duplication, and states exactly what the company wrote — with the source URL in the row to prove it.

Your slice vs a typical postings data vendor

Quantic AI datasetTypical postings vendor
Scoperole × geo × window you definetheir full catalog, filtered
Buyingself-serve, cents per runlicensing negotiation
Schemayour columnstheir taxonomy
Freshnesscollected at request timetheir crawl cycle
Sourcecareer pages, URL per rowaggregated, provenance varies

What does a weekly hiring scan cost?

With Quantic AI (planned): 120 postings × $0.03 ≈ $3.60 per weekly scan plus metered pipeline; refreshes of known career pages drop to $0.01 per record. On the raw APIs the same run meters at about $0.05 — code on you.

Get my free API key

FAQ

The short answers on pricing, formats and setup.

Something else? Ask us →

Where can I get job postings data?

Three routes: government statistics (BLS, Eurostat — aggregated, months behind), enterprise vendors licensing billion-row catalogs, or collecting the slice you need from public sources yourself. This page is the third route, automated: the Google Jobs vertical finds postings, career pages fill in the detail.

Is there a free job postings dataset?

Public statistics are free but aggregated — you can't see individual roles or companies. Here, $1 of usage renews free every month: on the raw APIs that's thousands of scraped pages, and at Quantic AI's planned record pricing it's a few dozen postings — enough to sample either path before paying.

Can I just use an open-source job scraper from GitHub?

You can, and for one or two boards it works. The cost shows up later: open-source job scrapers break whenever a board changes its markup, they ship without proxies so the big boards block you by volume, and nobody is on call when a nightly run returns empty. This runs the same idea as a service — search finds the postings, extraction reads them, the proxy network absorbs the blocking, and failed calls are never billed.

How do I scrape job postings with Python?

Call the API from Python instead of building the stack: one request to the search endpoint with the jobs vertical returns the postings, one scrape call per career page returns clean Markdown or the exact fields you name. That replaces requests plus BeautifulSoup plus a proxy pool plus retry logic. The same endpoints are also MCP tools, so a Python agent can drive them itself.

What's the difference between BLS statistics and raw postings data?

BLS-style statistics count openings by sector after the fact — clean, official, aggregated. Raw postings are the openings themselves: title, company, location, requirements, posted date. You need the raw rows for sales triggers, salary benchmarks, competitor hiring maps or skills analysis.

Can it scrape LinkedIn or Indeed?

No — walled boards behind logins are out of scope by design. The pipeline reads the Google Jobs vertical and the open web: company career pages, public ATS pages and boards that don't gate content. For hiring-signal analysis, career pages are the primary source anyway.

What fields does a postings dataset include?

Title, company, location, posted date, source URL and — where the posting states them — salary range, seniority, remote policy and required skills. You name the columns in the prompt; extraction returns exactly those fields as CSV or JSON.

How much does job postings data cost here?

With Quantic AI, planned pricing: $0.03 per delivered posting — a weekly 120-posting scan lands around $3.60 plus metered pipeline, refreshes at $0.01 per record. No license, no annual contract. Developers can meter the same scan on the raw APIs for about $0.05 in unit prices.

The hiring market, in your columns

Start with $1 free every month — or join the Quantic AI waitlist for the full prompt-to-dataset service.

Get my free API key
Get my free API key