Hiring.Cafe Jobs Scraper

Scrape job listings from Hiring.Cafe: an aggregator that indexes jobs directly from company career sites (Workday, Greenhouse, Lever, and others) and shows each posting exactly as the employer published it.

Every record carries the full nested payload the source publishes (85+ structured job attributes, the company record, geo-coordinates), a curated flat view for filtering and sorting, and: with Fetch full job descriptions on: the complete posting description.

Two ways to start

1 · Start from URLs

Paste any mix of:

• Search links: a Hiring.Cafe URL with ?searchState=… in it (copy one from the site after searching). Each link is walked forward page by page; a ?page=N in the link is honoured, so the run continues from that page.
• Single job links: any …/job/… URL. The full posting is returned.

If any URL is set, the filters in section 2 are ignored.

2 · Or search by filters

Input	What it does
Search keyword	Job title, skill, or company name
Location	Country or city; empty = worldwide
Workplace type	Remote / Hybrid / Onsite
Employment type	Full Time / Part Time / Contract / Internship
Seniority level	From No Prior Experience Required to Senior Level
Posted within (days)	Only recent postings (default 30, max 365)
Salary transparent only	Only jobs with publicly disclosed pay

Output fields

Always present: job and company identifiers, apply URL, the site's public job URL, and the raw nested payload (v5_processed_job_data, job_information, company record, _geoloc).

Flat view (top level, only when the source publishes the field): normalized job title, employment types, seniority, workplace type and formatted location, remote-worldwide flag, salary min/max with currency and frequency (the most comparable pair the posting publishes, yearly first), salary-transparency flag, posting date, minimum years of experience, required education degrees, certifications, technical tools, language requirements, visa sponsorship, security clearance, benefits (retirement, 401k matching, tuition, relocation, PTO, parental leave), schedule flags (shifts, weekends, holidays, on-call, overtime, four-day week), physical requirements, company size/founding/HQ/funding/stock, and the posting's on-site view and apply counts.

With Fetch full job descriptions on (default): the complete posting description as HTML and as clean plain text. The search listings do not include the description: it is read from each job's own page, which costs one page read (~200 KB of proxy traffic per job) and takes a few seconds per job. Turn the toggle off for a fast listing-only crawl; those records then carry all the structured attributes but no description text.

Resume & recurring monitoring

• Resume from a previous run: continue ONE interrupted run: jobs already collected there are skipped (and not re-charged).
• Incremental mode: for a schedule. The first run returns everything as NEW; later runs return only NEW / UPDATED / REAPPEARED jobs, with the changed fields named. State is kept per search configuration; use a State key to name or deliberately share a campaign.
• Emit unchanged / Emit expired: off by default; they return (and bill) extra rows.

Note: turning Fetch full job descriptions on or off changes the record shape, so the first run after toggling it reports previously-seen jobs as UPDATED once, then stabilises.

Connection

The site refuses datacenter IP ranges. A residential connection is configured by default and is what this actor is built and tested against; your own proxy URLs win when you provide them, and a backup connection pool is built in for plans without residential credit. A run on a non-residential connection warns in its log and may return no results.

Limits

Maximum jobs caps the run (0 = unlimited). Each results page carries up to ~100 jobs. A page that cannot be loaded after retries stops that search and is reported: the run never silently returns fewer results than it found.