LinkedIn Jobs Scraper Pro v2

Production-grade LinkedIn job scraper. Multi-keyword, multi-location, JSON-LD enrichment, parallel detail fetch, real typed fields, dedup, XLSX/CSV/JSON export.

---

What it does

Search LinkedIn by job title × location and return a clean, enriched dataset. Every row includes the basics (title, company, location, date, Easy Apply, URL) plus the fields that actually matter for recruiting, sourcing, and market research:

• Salary (salaryMin, salaryMax, salaryCurrency, salaryUnit, salaryRaw) — parsed from JSON-LD baseSalary or inline ranges
• Company enrichment (companyUrl, companySlug, companyLogoUrl) — from JSON-LD hiringOrganization or card HTML
• Job classification (industry, jobFunction, experienceLevel, employmentType)
• Applicants — applicantsCount is an integer (47), applicantsRaw is the original text ("47 applicants")
• Recency (postedDate, postedTimeAgo, isNew, badge)
• Full description (jobDescription) + optional descriptionHtml
• Direct apply flag (directApply, easyApply as booleans)
• Search provenance (searchKeyword, searchLocation, scrapedAt)

---

Pricing

$0.50 per 1,000 jobs (pay per result, no monthly minimum).

Compared to:

• valig/linkedin-jobs-scraper: $0.40/1k — cheaper per job, more fields, parallel fetch
• bebity/linkedin-jobs-scraper: $15–29.99/month flat — no per-job pricing, capped by month

---

How it works

1. Phase 1 — listing scrape (guest API): pulls cards from linkedin.com/jobs-guest/jobs/api/seeMoreJobPostings/search for every (title × location) combo. No login, no auth wall.
2. Dedup by jobId across the whole run.
3. Phase 2 — parallel detail fetch with asyncio.gather + asyncio.Semaphore(concurrency). Each detail page is parsed for JobPosting JSON-LD (most rich fields live here) and BeautifulSoup is used for fields JSON-LD doesn't carry (applicants, seniority, Easy Apply button).
4. Optional XLSX/CSV export to the run's Key-Value Store.
5. Push to dataset with strict typing: bools are bools, counts are ints, missing data is null.

---

Input

Field	Type	Default	Description
job_titles	string list	["Head of Engineering"]	Titles to search (required)
locations	string list	["United States"]	LinkedIn location strings
days_posted	integer	60	Filter by posting recency (1–365)
max_jobs_per_keyword	integer	200	Per-(title × location) cap, 8 pages max
seniority_levels	select	4,5,6	LinkedIn f_E codes
work_type	select	All	On-site / Remote / Hybrid / All
employment_types	select	All	f_JT codes
easy_apply_only	bool	false	Restrict to Easy Apply
fetch_job_details	bool	true	Visit each job page for full data
use_apify_proxy	bool	true	RESIDENTIAL proxy, per-request rotation on both listing and detail calls
concurrency	integer	10	Parallel detail-fetch workers (1–20)
listing_concurrency	integer	8	Parallel (title × location) combos, each gets its own proxy
min_seniority	integer	0	Post-fetch seniority filter (0–5, 0 = no filter)
max_results	integer	500	Cap on final output (after fit-score sort)
include_description_html	bool	false	Add raw HTML field (large)
min_salary	integer	0	Drop jobs below this salary (in source currency)
output_format	select	json	json / xlsx / csv / all

Example input

{
  "job_titles": ["Head of Engineering", "VP of Technology", "Delivery Director"],
  "locations": ["India", "United States", "United Kingdom"],
  "days_posted": 30,
  "max_jobs_per_keyword": 50,
  "seniority_levels": "5,6",
  "work_type": "2",
  "fetch_job_details": true,
  "use_apify_proxy": true,
  "concurrency": 8,
  "output_format": "all"
}

Example output

{
  "jobId": "4418924625",
  "title": "Senior Director, AI Product Operations & Solutions",
  "company": "Blue Machines AI",
  "companyUrl": "https://www.linkedin.com/company/blue-machines-ai",
  "companySlug": "blue-machines-ai",
  "companyLogoUrl": "https://media.licdn.com/.../blue_machines_ai_logo",
  "location": "Mumbai, Maharashtra, India",
  "postedDate": "2026-05-26T08:00:00.000Z",
  "postedTimeAgo": "1 day ago",
  "isNew": true,
  "badge": "Actively Hiring",
  "salaryMin": 5000000,
  "salaryMax": 8000000,
  "salaryCurrency": "₹",
  "salaryUnit": "year",
  "salaryRaw": "₹50,00,000 - ₹80,00,000 a year",
  "easyApply": false,
  "remoteHybrid": false,
  "employmentType": "Full-time",
  "experienceLevel": "Director",
  "jobFunction": "Information Technology",
  "industry": "Software Development",
  "applicantsCount": 47,
  "applicantsRaw": "47 applicants",
  "validThrough": "2026-07-26T00:00:00.000Z",
  "directApply": false,
  "jobDescription": "We are looking for a Senior Director...",
  "jobUrl": "https://in.linkedin.com/jobs/view/4418924625",
  "searchKeyword": "Delivery Director",
  "searchLocation": "India",
  "scrapedAt": "2026-06-23T10:15:00.000Z"
}

---

Output formats

• JSON — always pushed to the dataset. Pipe straight to apify-client, BigQuery, or a NoSQL store.
• XLSX — written to the run's Key-Value Store as jobs-<runId>.xlsx when output_format=xlsx or all. Opens in Excel/Sheets with auto-typed columns.
• CSV — written to the run's Key-Value Store as jobs-<runId>.csv when output_format=csv or all.

Performance

Run	Combos	Detail fetch	Est. time (concurrent=5)	Est. time (concurrent=10)
Quick scan, no details	10 × 2	off	~5 min	~5 min
Standard	10 × 2	on	~20 min	~12 min
Deep run	30 × 4	on	~2.5 hrs	~1.5 hrs

---

Limitations

• LinkedIn's guest API is unauthenticated. The card HTML is reliable; the detail HTML occasionally hits a soft auth wall (job is then pushed with jobDescription="" and the URL is preserved).
• Some fields (salary, applicants) are not on every posting — nulls are expected.
• LinkedIn rotates its HTML classes; this version pins the Q2 2026 class set and falls back to JSON-LD for resilience.

---

License & ToS

This actor is for personal use, research, and licensed commercial use. Respect LinkedIn's ToS and applicable data-protection laws (GDPR, etc.) when storing personal data.