AI & ML Job Listings Scraper

Scrape AI, ML, data science, and related engineering job listings from foorilla.com/hiring — the career platform that now hosts the job board formerly published at aijobs.net. Extracts salary range, seniority, years of experience, remote policy, skills, education requirements, tasks and the apply URL for each posting.

What data does it extract?

Each record includes:

Field	Description
job_id	Numeric job ID
job_slug	URL slug for the job
title	Job title
company_name	Hiring company, as rendered (the site truncates this in the job view, e.g. G...)
location_raw	Location string as shown; semicolon-separated when a posting lists several
location_city	City or region (first location when several are listed)
location_country	Country (first location when several are listed)
is_remote	Boolean — remote-eligible
remote_policy	remote / onsite
employment_type	full-time / part-time / contract / internship
seniority	entry / mid / senior / staff / principal
years_experience_raw	Years-of-experience hint as shown (e.g. ~10yoe)
salary_min	Minimum salary in salary_currency
salary_max	Maximum salary in salary_currency
salary_currency	Currency code (USD, EUR, GBP, SEK, PLN, CHF, …)
salary_raw	Raw salary string from the page
is_salary_estimated	Boolean — true when the site labels the range (estimate) rather than sourcing it from the employer
posted_at	Exact publication date (ISO YYYY-MM-DD)
role_category	Inferred: ai-engineer, ml-engineer, data-scientist, research, mlops, data-engineer, other
tech_tags	Comma-separated skill / tech-stack tags
education_requirements	Comma-separated education requirements
job_roles	Comma-separated role taxonomy the site assigns
tasks	Pipe-separated responsibilities
perks	Pipe-separated perks and benefits
views	View count the site reports
apply_clicks	Apply-click count the site reports
saves	Save count the site reports
language	Language code of the posting
apply_url	Direct apply link
profile_url	Canonical job page URL
scraped_at	ISO-8601 timestamp when the record was collected

How does it work?

The actor pages the public job listing at foorilla.com/hiring/jobs/ to discover postings, then reads each job detail page and returns every field above — salary, seniority, skills, education and tasks — as one flat record per job, ready to load straight into a sheet or a database.

The maxItems input controls how many records to collect.

Coverage

This actor returns the freshest ~105 public postings, not the entire board. foorilla serves anonymous callers listing pages 1 and 2 (roughly 55 + 50 jobs) and answers page 3 onward with "Sign in to view more". There is no unauthenticated surface that reaches deeper — verified 2026-09-05 against paged and filtered listings, the "top" listing (which ignores paging), sitemap.xml, sitemap-jobs.xml, RSS and /api/jobs/, none of which expose the rest of the inventory.

Setting maxItems: 0 collects everything reachable, which is that ~105, not the full board.

Historical note: until 2026-09-03 this actor scraped aijobs.net via a ~48,000-URL sitemap. That domain now redirects to foorilla.com and the sitemap is gone. The much smaller number above reflects what the successor site exposes publicly, not a regression in the actor.

Input

{
  "maxItems": 100
}

Parameter	Type	Default	Description
maxItems	integer	10	Maximum number of job records to collect. Set to 0 for everything reachable.

Use cases

• AI talent market research — track demand across roles (AI engineer, ML engineer, MLOps, research scientist) over time
• Salary benchmarking — compare compensation by role, seniority, and region, separating employer-published ranges from site estimates via is_salary_estimated
• RecOps & ATS enrichment — bulk-import active job listings for sourcing workflows
• Remote work signal monitoring — filter by remote_policy for location-agnostic hiring intelligence
• Tech stack trend analysis — aggregate tech_tags to identify the fastest-rising skills

Notes

• Salary values are annualized and reported in the posting's own currency; salary_currency carries the code and salary_raw the original string. Many ranges are the site's own estimates — check is_salary_estimated before treating one as employer-published.
• posted_at is the exact publication date the site prints, not an estimate derived from a "posted N days ago" label.
• company_name is truncated by the site in this view. The full name is only on the company index, which the site's robots.txt disallows, so it is emitted as rendered rather than guessed.
• Because coverage is recency-ordered, repeated runs are the intended way to accumulate history — each run returns the newest postings.