LinkedIn Jobs Scraper

---

🎯 What this scrapes

Point this Actor at one or more job-title keywords and one or more locations, and it runs the full cross product as independent searches against LinkedIn's public guest jobs search. Each search unit walks the result pages until it reaches the end of that search's job list or a safety cap you set. Every job card is parsed into a typed row — title, company name and LinkedIn URL, the job's own location, a cleaned job URL, posting date, and an optional benefit/status label. Multiple keyword/location pairs run in the same batch, and one pair's error never sinks the others.

🔥 What we handle for you

• 🛡️ Browser fingerprint rotation — curl-cffi replays a real browser's TLS handshake (Chrome and Firefox profiles), rotated per request.
• 🔁 Retries with exponential backoff on 408 / 429 / 503 and network errors — up to 5 attempts, Retry-After honoured.
• 🧱 Per-search fault isolation — a failed or empty keyword/location pair is logged and skipped; every other pair in the batch still runs.
• 🧊 Clean, typed dataset rows — Pydantic-validated, deduplicated by job id, ISO-8601 timestamps, JSON / CSV / Excel export straight from the Apify Console.
• 💰 Pay-Per-Event pricing — you only pay for job rows that land in your dataset. No data, no charge.

💡 Use cases

• Talent sourcing — pull fresh postings for a role/location combo to see which companies are hiring right now.
• Market and salary research — batch a role across multiple cities or countries to gauge hiring volume and posting cadence.
• Recruiting agency lead gen — find companies actively posting a given role as warm outreach targets.
• Job board aggregation — feed a clean, typed dataset of LinkedIn postings into your own job board or alert pipeline.

⚙️ How to use it

1. Click Try for free at the top of the page.
2. Add one or more job-title keywords to keywords; add one or more locations to locations — most other fields have sensible defaults.
3. Click Start. Rows stream into the run's dataset as each keyword/location pair is searched.
4. Export from Storage → Dataset as JSON, CSV, or Excel — or fetch via the API.

📥 Input

Field	Type	Required	Default	Notes
keywords	array	yes	["python developer", "data engineer"]	Job-title/keyword searches — one guest jobs search per item, run against every location.
locations	array	no	["United States"]	Locations to search. The Actor runs the full keywords x locations cross product.
maxPagesPerSearch	integer	no	10	Safety cap on the offset-walk depth (10 job cards per page) for one search unit.
maxJobsPerSearch	integer	no	100	Cap on job rows collected for a single search unit before its walk stops early.
maxTotalResults	integer	no	500	Cap on total job rows emitted across the whole run.
maxRunSeconds	integer	no	600	Wall-clock budget across all search units; the run ships whatever it collected so far. Capped at 3 480s, below the platform's own hard limit.
proxyConfiguration	object	no	{"useApifyProxy": true}	Apify Proxy config — defaults to a bare flag (DATACENTER); every probe of this specific endpoint cleared direct and unproxied.

Example input

{
  "keywords": ["python developer", "data engineer"],
  "locations": ["United States", "United Kingdom"],
  "maxJobsPerSearch": 100,
  "maxTotalResults": 500,
  "maxRunSeconds": 600,
  "proxyConfiguration": {
    "useApifyProxy": true
  }
}

📤 Output

Every row is one parsed job posting.

Field	Type	Notes
search_keyword	string	The input keyword searched (echoed, not re-parsed).
search_location	string	The input location searched (echoed, not re-parsed).
job_id	string	Numeric job id parsed from the card's data-entity-urn.
title	string	Job title.
company_name	string | null	Hiring company name.
company_url	string | null	Company's LinkedIn URL, when present.
location	string | null	The job's own location string (may differ from the search's input location).
job_url	string | null	Cleaned canonical job URL — tracking query params stripped.
posted_at	string | null	ISO calendar date (YYYY-MM-DD) the job was listed.
posted_text	string | null	Human-readable posting age, e.g. "2 months ago".
benefit_label	string | null	Optional benefit/status label LinkedIn shows on the card (e.g. "Actively Hiring").
page_number	integer	1-based page within this search unit's offset walk.
scraped_at	string	ISO-8601 UTC timestamp at parse time.

Example output

{
  "search_keyword": "data engineer",
  "search_location": "United States",
  "job_id": "3812345678",
  "title": "Senior Data Engineer",
  "company_name": "Acme Corp",
  "company_url": "https://www.linkedin.com/company/acme-corp",
  "location": "Austin, TX",
  "job_url": "https://www.linkedin.com/jobs/view/3812345678",
  "posted_at": "2026-09-10",
  "posted_text": "1 week ago",
  "benefit_label": "Actively Hiring",
  "page_number": 1,
  "scraped_at": "2026-09-17T14:03:22.104Z"
}

💰 Pricing

Pay-Per-Event — you pay only when these events fire:

Event	USD	What it is
actor-start	$0.20	Base fee, once per run (warm-up, proxy resolution)
job-result	$0.003	Per job-posting row written to the dataset

Example: 1 000 job rows in one run ≈ $3.20 ($0.20 + 1 000 × $0.003). No subscription, no minimum — Apify gives every new account free trial credit.

🚧 Limitations

• Real page size is 10 job cards per response, not 50 — size maxJobsPerSearch / maxPagesPerSearch with that in mind.
• Default proxy tier is DATACENTER, not RESIDENTIAL — a bare {"useApifyProxy": true}. Every probe of this specific guest endpoint (both browser profiles, multiple pagination offsets, both stop-condition signals) cleared direct and unproxied with no anti-bot signal observed. Add apifyProxyGroups: ["RESIDENTIAL"] yourself if you ever see blocking on your account.
• A broad or nonsense keyword can still return generic matches instead of zero results. That's LinkedIn's own guest-search fallback behavior — this Actor doesn't add or remove client-side relevance filtering, so a very loose search may return jobs you didn't expect.
• When a search unit's walk stops early, the run's status message is prefixed [TRUNCATED] instead of [COMPLETE], so you always know whether you got everything or hit a cap.
• No login or session persistence — every request is anonymous, reading only what LinkedIn's guest jobs search already serves to a public visitor.

❓ FAQ

Is this legal?

We only fetch what LinkedIn's public guest jobs search already serves to anonymous visitors — no login, no session cookies, no credential harvesting. Respect LinkedIn's terms of service before using output commercially.

What if a keyword/location pair has no jobs?

The run still succeeds. That pair's rows are simply zero, and the status message says so — a real zero-match result is never treated as a failure.

Why does my run say [TRUNCATED] instead of [COMPLETE]?

[TRUNCATED] means at least one search unit's walk stopped before reaching the end of its result list — a safety cap (maxJobsPerSearch, maxPagesPerSearch, maxTotalResults, or maxRunSeconds) was hit first. This Actor bills per row, so we never let a partial run's status message read the same as a finished one. Raise the relevant cap and re-run for the rest.

Why are some fields null?

LinkedIn's guest-search markup doesn't guarantee every field on every card — for example, not every posting carries a benefit/status label. We never fabricate a value — unconfirmed fields come back null rather than a guess.

Can I track new postings over time?

Each run is a single snapshot. Schedule the Actor daily/weekly in Apify Console and diff successive dataset exports to catch new listings.

💬 Your feedback

Spotted a bug, hit a weird edge case, or need a new field? Open an issue on the Actor's Issues tab on Apify Console — we ship fixes weekly and we read every report.

---