LinkedIn Jobs Scraper | AI Enrichment

Get LinkedIn jobs ready for alerts, job boards, and spreadsheets. Search by role and location, keep full descriptions and original job links, and skip jobs already returned by earlier runs. Add AI enrichment or English translation when you need them.

Why choose this Actor?

• Less cleanup: consistent fields for titles, companies, locations, salaries, skills, and application links, when available.
• Fewer repeat alerts: built-in deduplication remembers jobs already delivered for the same search.
• More useful details: optional AI reads descriptions to fill missing facts without replacing source data.
• Ready to connect: export results or use the API, n8n, Make, Airtable, and MCP.

Start with a small search

1. Enter a Job keyword and Location.
2. Set Maximum results to 5 and leave First-run mode and Add company details off for a basic search.
3. Set Apify's Maximum cost per run to $0.10, then click Start.
4. Open the results to review the jobs and export your data.

AI enrichment and translation are optional paid extras. The cost cap can reduce or stop output; the Pricing tab is authoritative for current charges.

Want to try the extras? Enable First-run mode for up to five jobs with Silver AI enrichment and selected-field English translation. It keeps your search and filters, disables deduplication and analytics, and omits descriptions from the output. Turn it off again to use your own settings.

Choose your results

Setting	Default / option	What it does
Job keyword / Location	—	Search for a role or skill in a city, region, or country. An empty location searches worldwide.
Posted within	—	Choose how recent the jobs should be. Default: past 24 hours.
Work arrangements	—	Select remote, hybrid, on-site, or any combination. Leave empty for all.
maxItems	100	Maximum jobs returned; up to 1,000 per run. 0 requests that limit.
Result order	—	Newest or oldest first.
Cross-run deduplication	—	Skip jobs already delivered. Turn off for repeatable one-off searches.
Include descriptions	—	Keep full text and HTML in raw. Turn off for smaller exports.

For separate alerts, use separate deduplication keys. An empty key keeps history separate for each search; reusing a named key shares history across those searches within your account. Duplicate tracking is bounded, so it is not permanent history.

Advanced controls support up to eight searches per run, job-field filters, strict location checks, and company filters. Use the examples in the input form's JSON editors. Clear the main keyword and location when providing multiple searches.

Optional extras

• AI enrichment: fills supported missing fields from the job description. Silver is the standard tier; Gold adds a second check. Use Silver until Gold appears in the current Pricing table. AI can make mistakes; review important facts.
• English translation: translates selected job fields, including titles, requirements, and benefits. Full descriptions, company names, locations, skills, and URLs stay in their original form.
• Company details: adds available facts from the job's linked public company page, with no separate result-event charge. Company filters require this option. Off by default. Unavailable profiles are excluded by default when company filters are used.

What you get

Each job has its LinkedIn ID and URL, structured job details, optional company facts and AI status, and the full description when requested. Missing facts remain unknown; the Actor does not guarantee that every field will be filled.

The JSON format is nomad-agent-job-v1: identity, data, custom, llm, raw, and schemaVersion. null means unknown; [] means confirmed empty. A matching workplace search can still return an unknown per-job arrangement. Sanitize source HTML before displaying it.

Use the API or an integration

from decimal import Decimal
from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor(
    "job-atlas/linkedin-enrich-translate-normalize-scraper"
).call(run_input={
    "schemaVersion": "nomad-agent-job-search-input-v1",
    "keyword": "software engineer",
    "location": "Spain",
    "maxItems": 5,
    "firstRunMode": False,
    "aiEnrichment": {"enabled": False, "accuracy": "silver"},
    "translateToEnglish": False,
    "companyProfileEnrichment": False,
    "analyticsEnabled": False,
    "dedupe": {"enabled": False, "key": ""},
}, build="latest", max_items=5, max_total_charge_usd=Decimal("0.10"))

if run["status"] != "SUCCEEDED" or not run.get("buildId") or not run.get("buildNumber"):
    raise RuntimeError("Actor run failed or omitted its resolved build identity")

for job in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(job["data"]["title"], job["identity"]["url"])

This standalone example follows latest. Check the current default build API for the default build number and immutable ID. Maintained integration templates use latest; follow the selected template's compatibility guide when running it. A 1.0.x check alone does not prove compatibility.

Integration guides and templates cover MCP, n8n, Make, Airtable, and agent skills.

Pricing and limits

You pay per returned job, plus applicable translation and successful AI-enrichment charges. Check the Pricing tab for current rates and scheduled changes.

Results depend on available public LinkedIn postings. Filters, source restrictions, and your budget can reduce the count. Large searches take longer. Check RUN-SUMMARY for the returned count, limits, and any retry suggestion; check llm.status for each job's AI outcome. Strict location and company filters can exclude jobs with missing evidence.

This independent Actor is not affiliated with or endorsed by LinkedIn. Contact the Actor creator through the Apify issue tab for help or privacy requests. For corrections or removal, include the job URL or ID and avoid private information.

Job Atlas

Explore the Job Atlas job-data and matching Actors. Use latest and retain the immutable build ID returned by each run.

• Linkedin
• Euraxess
• YC
• Scorer

Website and integration guides | Source and client examples