LinkedIn Jobs Scraper Pro
Pricing
Pay per usage
Search LinkedIn jobs by title, location, seniority & work type. Multi-location, remote filter, Easy Apply detection. Export to XLSX, CSV, JSON.
Pricing
Pay per usage
Rating
0.0
(0)
Developer
Sonal Sharma
Maintained by CommunityActor stats
0
Bookmarked
17
Total users
1
Monthly active users
17 days ago
Last modified
Categories
Share
All notable changes to LinkedIn Jobs Scraper Pro.
scrape_keyword / fetch_search_page now call proxy_cfg.new_url() per request, not per run. Previously a single shared proxy_url was baked into one httpx.AsyncClient for all 88+ (title × location) calls, so LinkedIn was throttling from a fixed IP. Detail phase already rotated correctly; this brings listing to parity.scrape_keyword no longer bails out when a page returns < 25 cards. Lets the scraper page through LinkedIn's fuzzy guest-API pagination instead of stopping at the first partial page.seniority_levels: "4,5,6" → "" (all levels — LinkedIn's f_E filter is no longer applied by default, which was the single biggest volume bottleneck)max_jobs_per_keyword: 50 → 500 (the schema max — 20 pages)days_posted: 30 → 90 (full quarter)concurrency: 5 → 10listing_concurrency: 5 → 8min_seniority: 4 → 0 (no post-fetch filter)max_results: 100 → 1000 (the schema max)listing_concurrency is now in the input schema (was only available via API before)JobPosting extraction from detail pages → unlocks salaryMin, salaryMax, salaryCurrency, salaryUnit, salaryRaw, industry, jobFunction, validThrough, directApply, descriptionHtml, structured company info$120k–$180k a year, ₹50–80 LPA, etc.)companyUrl, companySlug, companyLogoUrl, postedTimeAgo, isNew, badgeconcurrency input (1–20) for parallel detail fetching — runs 3–5× fasterinclude_description_html input (default off) — adds raw HTML to outputmin_salary input — filter by parsed minimum salaryoutput_format input — json (dataset), xlsx (KV store), csv (KV store), or allopenpyxleasyApply, remoteHybrid, isNew, directApply are now bool; applicantsCount is int|null"Over 200 applicants", "47 applicants", "1,200 applicants")actor.jsonasyncio.gather + asyncio.Semaphore)RESIDENTIAL per request, not per runjobId (or cleaned URL) — survives across keywords and locationsget_job_details was never invoked)data-delayed-url is preferred over src for company logos (skips placeholder)concurrencyActor.config is a dataclass, not a dict — runtime was crashing on actorRunId accessexperienceRequirements from JSON-LD is in months, not a label — fall back to the visible seniority string from description__job-criteria-itemsameAs from JSON-LD is sometimes a list, not a string — handled both shapesemploymentType, industry, jobFunction from JSON-LD may be a list or dict — flatten to stringdescription__job-criteria-item parsing now runs on every detail fetch, not only when JSON-LD is missingOriginal linkedin-jobs-scraper-pro (build 0.0.29). Guest API listing + sequential httpx detail fetch. Kept the architecture; this version is a full rewrite of the parsing and orchestration layer.