Bayt Jobs Scraper (MENA)

Scrapes job listings from bayt.com — the largest job board across the Middle East and North Africa — over plain HTTP. No browser, no login.

Two-step scrape, because Bayt's search page alone isn't enough

Bayt's search results carry a JSON-LD ItemList, but it's just 30 URLs — no salary, no company, no description. The full record lives on each job's own page, as a proper schema.org JobPosting block. This Actor fetches both: the search page for the URL list, then each job page for the complete record.

That means one listing costs two HTTP fetches, not one — reflected honestly in the per-job charge, since a two-step scrape has real cost the pricing should not hide.

No free-text search — Bayt routes by URL path

There's no ?q= parameter here. Bayt organizes search through URL paths like en/uae/jobs/python-jobs. To use this Actor, browse bayt.com, run the search you want, and copy the path segment after bayt.com/:

{ "categoryPaths": ["en/uae/jobs/python-jobs", "en/saudi-arabia/jobs/accounting-jobs"] }

What you get per job

{
  "title": "Senior Backend Engineer – Node.js",
  "company": { "name": "Black Pearl", "profile_url": "https://www.bayt.com/en/company/black-pearl-1941858/" },
  "location": { "raw": "United Arab Emirates", "region": "United Arab Emirates", "country_code": "AE" },
  "employment_type": "FULL_TIME",
  "posted_at": "2026-08-12T17:00:00Z",
  "valid_through": "2026-10-12T00:00:00Z",
  "description_html": "..."
}

location.country_code is a proper ISO 3166-1 alpha-2 code — Bayt's own addressCountry field arrives as a bare code like "AE" rather than a full country name, and this Actor detects that shape rather than assuming one or the other.

Salary is (almost) never there

salary is populated when Bayt's schema.org data discloses it, but in practice baseSalary is null on effectively every listing in this market — MENA job ads rarely publish pay. Don't build a pay-benchmark pipeline expecting otherwise; use this Actor for listing volume, company activity and description mining instead.

Input

Field	Type	Notes
categoryPaths	array	Bayt URL path segments, e.g. en/uae/jobs/python-jobs.
maxItems	integer	Per path. 0 = unlimited, bounded by Bayt's pagination depth.
incremental	boolean	Only listings not seen in previous runs.
proxyConfiguration	object	Datacenter is normally enough — no anti-bot layer encountered.

Known limits

• Two fetches per row. Slower and costs more per item than a single-page scrape, in exchange for a complete record instead of a title and a link.
• No numeric salary in practice. See above.
• No free-text search. You navigate by category path, not by keyword.

Local development

pip install -r requirements.txt
python test_local.py "en/uae/jobs/python-jobs" --max 20 --out sample_output.json
python test_local.py "en/saudi-arabia/jobs/accounting-jobs" --max 10

sample_output.json in this folder is real output from a live run, kept so the schema can be reviewed without running anything.