LinkedIn Jobs Scraper: Beyond the 1000 Limit, No Login

Get LinkedIn job postings for any keyword and place, for a list of companies, or for a LinkedIn job search URL. No login, no cookies, no LinkedIn account: only the pages LinkedIn shows to visitors who are not signed in.

LinkedIn hands out about 1,000 jobs per search, however many it counts. This scraper divides a larger search by state or region, city, company and date, reads each part, and merges the results without duplicates. Every run reports what it found against LinkedIn's own count.

It collects job and company data only. Recruiter and "hiring team" names and profile links are never collected.

What you can do with it

• Follow hiring for a job title in a country or city: schedule a run with Date posted set to the last 24 hours.
• Watch what a list of companies is hiring for with the Companies input.
• Collect a large set of postings for a keyword, past the 1,000 a single LinkedIn search hands out.
• Skills and seniority analysis on full descriptions, seniorityLevel and employmentType (details optional).

What you get

One record per job, never the same job twice in a search:

• Job: jobId, title, url, location, postedDate, postedText ("2 days ago"), badge (LinkedIn's note on the card, such as "Actively Hiring" or "Be an early applicant")
• Company: companyName, companyUrl (its LinkedIn page)
• Pay, when LinkedIn shows it: salaryText (for example $90,000.00/yr - $140,000.00/yr). On the result list alone almost no job shows pay; with details on, 24% of jobs in our 500-job test run did.
• Workplace: workplaceType (Remote, Hybrid or On-site) only when the job's title or location says so, which is rare (1% to 4% of jobs in our runs). LinkedIn does not show visitors its own workplace label.
• With details switched on: description (full text, plain), seniorityLevel, employmentType, industries, applicantsText ("Over 200 applicants"), applyType (Easy Apply or Company website), applyUrl, companyId, acceptingApplications, detailsFetched
• Other: search (which keyword or URL produced it), scrapedAt

{
    "jobId": "4444613232",
    "title": "Enterprise Data Analyst",
    "companyName": "Husch Blackwell",
    "companyUrl": "https://www.linkedin.com/company/husch-blackwell",
    "location": "Austin, TX",
    "postedDate": "2026-09-26",
    "postedText": "1 week ago",
    "salaryText": null,
    "badge": null,
    "workplaceType": null,
    "url": "https://www.linkedin.com/jobs/view/enterprise-data-analyst-at-husch-blackwell-4444613232",
    "applyUrl": "https://www.linkedin.com/jobs/view/enterprise-data-analyst-at-husch-blackwell-4444613232",
    "applyType": "Company website",
    "description": "Husch Blackwell LLP is a full-service litigation and business law firm with multiple locations across the United States...",
    "seniorityLevel": "Not Applicable",
    "employmentType": "Full-time",
    "jobFunction": null,
    "industries": "Law Practice",
    "applicantsText": "Over 200 applicants",
    "companyId": "162725",
    "acceptingApplications": true,
    "detailsFetched": true,
    "search": "data analyst in Austin, Texas, United States",
    "scrapedAt": "2026-10-08T16:28:55.614Z"
}

Input

• Keywords — one search per line. Leave empty to get every job in a place or at the companies you list. If you give no keyword, location, company or search URL (for example an API call with an empty input), the example search data analyst in Austin, Texas, United States runs.
• Location — a country, a state or region, or a city, written as on LinkedIn: United States, Texas, United States, Austin, Texas, United States. Empty searches worldwide.
• Date posted — last hour, 24 hours, 3, 7, 14 or 30 days.
• Distance — miles around a city (0 is the city itself).
• Companies — LinkedIn company page addresses, the last part of one (microsoft), or numeric IDs.
• Title must contain / must not contain — keeps or drops jobs by words in the title. Jobs dropped are not charged.
• Job type, Experience level — see "Filters LinkedIn no longer applies" below.
• Easy Apply only, Under 10 applicants only
• LinkedIn job search URLs — paste the address of a linkedin.com/jobs/search results page.
• Maximum jobs per keyword or URL — default 100. You are never charged for more.
• Go past LinkedIn's 1,000-job limit — on by default; only matters when your maximum is above 1,000.
• Stop at LinkedIn's count — off by default; see below.
• Include job details and full description — off by default.

{
    "keywords": ["data engineer"],
    "location": "United States",
    "datePosted": "604800",
    "titleIncludes": ["data engineer"],
    "maxJobsPerSearch": 5000,
    "includeDescriptions": true
}

FAQ

How do I get more than 1,000 jobs from a LinkedIn job search?

Set Maximum jobs per keyword or URL above 1,000 and leave Go past LinkedIn's 1,000-job limit on. The scraper divides the search by state or region, city, company and date, reads each part and merges the results without duplicates. Jobs from the narrower searches are increasingly loose matches; use Title must contain for a tight list. The measured runs are under "What we measured".

Does it need a LinkedIn account or cookies?

No. It reads only the pages LinkedIn shows to visitors who are not signed in.

Can I filter LinkedIn jobs by remote, salary or sort order?

No. LinkedIn ignores those filters for visitors who are not signed in and shows nothing to check them against. To look for remote jobs, put "remote" in your keywords and use Title must contain. Job type and experience level are checked on each posting's own page; date posted, company, distance, Easy Apply and under 10 applicants are applied by LinkedIn.

Does it collect recruiter names or profiles?

No. It collects job posting and company data only. Recruiter and "hiring team" names and profile links are never collected.

How much does it cost?

$0.50 per 1,000 jobs on Apify's Free plan, plus $0.50 per 1,000 jobs with details when you ask for details or use the job type or experience filters, and a start fee of $0.00005 per run. Paid Apify plans pay less: Bronze 10%, Silver 20% and Gold 30% less. Runs that fail or return nothing cost nothing.

How many jobs you can get: read this first

Three facts about LinkedIn's public job search, all measured in October 2026:

1. A search hands out about 1,000 jobs. Requests past position 1,000 are refused, and many searches stop earlier (we saw 860 to 1,000).
2. LinkedIn's count for a search stops near 12,000. "Data engineer", "nurse" and "project manager in California" all count about 11,900 in the United States, although far more jobs match.
3. A narrower search returns jobs the wider one leaves out. The 50 states searched one by one return several times the jobs LinkedIn counts for the whole country. The extra jobs are increasingly loose matches: in a search for "data engineer", 97% of the first 1,000 jobs have "data engineer" in the title, against 33% of the first 12,000.

So the scraper reads your search as LinkedIn lists it (the best-ranked 1,000), then divides it: a worldwide search by country, a country by state or region, a state by city, and whatever is still too large by the companies LinkedIn offers as a filter and by date posted. The places come from the locations of the jobs already found. All parts are read side by side from their best-ranked jobs down, so whatever your maximum is, you get the top of every part. Jobs from a narrower search are dropped if their location is not in the place that search was for, and the date filter is checked on every job.

A search ends when your maximum is reached, or when the narrower searches stop bringing new jobs. With Stop at LinkedIn's count on, it ends once as many jobs are found as LinkedIn counts.

What we measured

Seven searches, each with the maximum out of the way. Run on 8 October 2026.

With Stop at LinkedIn's count on:

Search	LinkedIn's count	Jobs saved	Requests	Run time	Check 1: best-ranked jobs found again	Check 2: recent matching-title jobs found
actuary, Columbus, Ohio	365	392	60	7 s	69% of 305	12 of 12
data analyst, Austin, Texas	2,860	2,471	703	53 s	96% of 801	3 of 3
registered nurse, Chicago, Illinois	3,577	3,645	344	22 s	98% of 839	84 of 145 (58%)
marketing manager, Germany	6,175	6,277	1,264	142 s	94% of 918	111 of 122 (91%)
software engineer, United Kingdom	11,723	11,122	1,840	178 s	99.8% of 995	26 of 26
data engineer, United States	11,937	12,418	1,256	126 s	94% of 1,000	33 of 40 (83%)
nurse, United States	11,940	12,058	1,142	102 s	89% of 989	74 of 602 (12%)

With the default settings (no stop at LinkedIn's count):

Search	LinkedIn's count	Jobs saved	Requests	Run time	Check 1	Check 2
data analyst, Austin, Texas	2,786	2,431	500	47 s	96% of 783	3 of 3
registered nurse, Chicago, Illinois	3,577	5,545	738	63 s	99% of 830	121 of 135 (90%)
marketing manager, Germany	6,174	6,975	1,433	133 s	95% of 957	112 of 122 (92%)
data engineer, United States	11,937	92,957	18,437	26 min	100% of 1,000	40 of 40

The actuary and United Kingdom searches end the same way in both settings. "Nurse" in the United States was not run with the default settings.

How to read the two checks. Neither depends on LinkedIn's count, which is not a true total.

• Check 1 takes the jobs LinkedIn's own list for the search handed out (up to 1,000, its best-ranked) and asks how many of them the divided searches found again by themselves. It says how well dividing reaches the jobs that match your keywords best. Every run computes it and stores it in the COVERAGE record.
• Check 2 is stricter. Just before each run we fetched a complete list of the jobs posted in the last hour (countries) or day (cities) for the same search, and counted how many of those with the keyword in their title were in the big run. It says how much of everything that matches a run holds, including jobs far down LinkedIn's ranking.

What the numbers say:

• With the default settings, 90% or more of the recent jobs with the keyword in their title were in the run in the four searches measured, at the price of many loosely related jobs in the largest one: of the 92,957 jobs saved for "data engineer", 5% have "data engineer" in the title.
• Stopping at LinkedIn's count gives a shorter, better-ranked list, and misses more. It held 83% to 100% of the recent matching-title jobs for specific job titles, 58% for nurses in Chicago and 12% for "nurse" in the whole United States, a search with far more than 12,000 matching jobs (about 600 with "nurse" in the title were posted in a single hour). For keywords that broad, leave the setting off and raise your maximum, or run one search per state.
• A big city is the weak case for the count. A city cannot be divided by place, only by its surrounding towns, companies and date. "Data analyst" in Austin gave 2,431 of the 2,786 LinkedIn counts (87%).
• More jobs than LinkedIn counts is normal (see fact 3). If you want only close matches from a large run, use Title must contain.
• A search under 1,000 jobs can come up short on a plain read, because LinkedIn's pages repeat some jobs and skip others. The scraper then divides it anyway, which is why the actuary run saved more than a plain read's 305.

Filters LinkedIn no longer applies

Since LinkedIn moved its public job search to "AI-powered search", it ignores the job type, experience level, workplace type, salary and sort-order filters for visitors who are not signed in: the same jobs come back with and without them (we tested each one).

What this scraper does about it:

• Job type and experience level are checked on each posting's own page. Every job returned states one of the types or levels you chose. This needs the posting's page, so jobs returned with these filters are charged as jobs with details, and runs take longer when few jobs pass. Note that about 80% of postings state "Not Applicable" as seniority level; those never pass an experience filter.
• Workplace type, salary and sort order cannot be filtered, because LinkedIn shows visitors nothing to check them against. To look for remote jobs, put "remote" in your keywords and use Title must contain. A pasted URL that carries these filters runs without them, and the log says so.
• Date posted, company, distance, Easy Apply and under 10 applicants are still applied by LinkedIn. In our test runs every job returned satisfied them (see "Checked on real runs").

Job details and descriptions

With Include job details and full description on, each posting's page is read. In our 500-job test run all 500 came back with a description, seniority level and employment type; 99% with industries and a company ID; 24% with pay. A job whose page cannot be read is still returned, with detailsFetched: false.

LinkedIn does not show visitors the employer's own application address, so applyUrl is the LinkedIn posting, where the Apply button is; applyType says whether that button applies on LinkedIn (Easy Apply) or leads to the company's website. jobFunction is kept for the day LinkedIn shows it again; today it is empty.

Speed

From our runs on Apify, 16 requests at a time:

• 100 jobs (the default input): 4 seconds.
• 12,000 jobs without details: about 2 minutes (100 to 180 seconds across four runs).
• 48,000 jobs without details: about 10 minutes; 93,000 jobs: 26 minutes.
• 500 jobs with details: 50 seconds; 800 with details: 53 to 74 seconds.

Pricing

You pay per job saved, and per job with details when you ask for details or use the job type or experience filters. Duplicates are never saved, so never charged. Jobs dropped by a filter are not charged. Runs that fail or return nothing cost nothing.

Reliability

Requests are made the way Chrome makes them and go through datacenter proxies, changing address after every refusal. A request LinkedIn refuses four times from datacenter addresses is retried through residential proxies; if half of recent requests are refused, the run uses residential proxies for two minutes before going back. LinkedIn's sign-in page, rate-limit answers (429, 999) and empty answers in the middle of a search are recognised and retried.

Measured: with 12, 16 and 24 requests at a time LinkedIn refused 1% to 5% of requests, which were repeated from another address; at 48 it refused 8%. The scraper runs 16 at a time. None of the 85 runs made while building this scraper failed.

If a search cannot be opened at all, the run says so in its status message and still returns the jobs of the other searches.

Checked on real runs

• Ten searches of 100 jobs in ten countries (United States, United Kingdom, Germany, Canada, Australia, France, India, United Arab Emirates, Brazil, Japan): all ten returned 100 jobs.
• Each filter once, with every returned job checked against it: date posted, last 24 hours (300 of 300 jobs inside the window) and last 7 days (300 of 300); distance 0 (200 of 200 in the city itself); companies (200 of 200 at the two companies asked for); under 10 applicants (100 of 100 carry LinkedIn's early-applicant note); Easy Apply (99 of 100 pages say Easy Apply, one does not say); job type (40 of 40); experience level (40 of 40); title must contain and must not contain (1,500 of 1,500); a pasted URL with a date and a job type filter (60 of 60).

Limits

• LinkedIn's public search is loose: beyond the best-ranked results, jobs are only related to your keywords. Use Title must contain for a tight list.
• Coverage of very broad searches in large countries, and of large single cities, is partial; see the numbers above.
• Workplace type, salary and sort order cannot be filtered; the employer's own application address and LinkedIn's workplace label are not available without signing in.
• Pay is missing where LinkedIn shows none (most postings).
• postedDate is a date without a time.
• Counts are per search; the same job can appear under two keywords.
• LinkedIn changes its public pages from time to time. Every run stores a COVERAGE record, so a drop shows up at once.

Notes

This scraper collects job posting and company data only, from pages LinkedIn shows to everyone. It does not log in and does not collect names, profiles or contact details of recruiters or other people.

Each run stores a COVERAGE record in its key-value store: per search, LinkedIn's count, jobs found and saved, the check on the divided searches, requests made, why it stopped, and how many new jobs each kind of narrowing brought.

Related scrapers

• Coles Australia Grocery Price Scraper: products, prices and specials from coles.com.au
• AliExpress Product Search Scraper: search results by keyword with prices, discounts and units sold
• eBay Active Listings & Auctions Search Scraper: active listings and auctions on five eBay sites
• Apple App Store Scraper: iOS top charts, search results and app details
• Zillow ZIP Code Search Scraper: homes for sale, rentals and recently sold homes
• Tripadvisor Places Scraper: hotels, restaurants and things to do for any place
• Google Maps Places Scraper: business listings across a whole city, postcode or radius
• Indeed Jobs Scraper: job listings on 12 Indeed country sites, descriptions optional
• Glassdoor Jobs Scraper: job listings with salary ranges and employer ratings
• Google Trends Scraper: interest over time, interest by region, rising queries and trending searches