LinkedIn Profile Lookup

Independent LinkedIn data extraction Actor. LinkedIn Profile Lookup is not affiliated with, endorsed by, or operated by LinkedIn or Microsoft. It reads information exposed on a public member page; access, visibility, field availability, and values can change without notice.

LinkedIn Profile Lookup turns one linkedin.com/in/... URL into a source-dependent person snapshot. A populated row contains 19 top-level keys covering provenance, identity, headline, About text, displayed location, image URL, follower and connection counts, source-inferred current companies and titles, experience, education, article samples, and activity samples.

• A verified Andrew Ng run returned one row with all 19 top-level keys, including 8 experience entries and 3 education entries.
• That same snapshot contained 2 source-visible article records and 10 activity records.
• Three later runs against public member URLs ended normally with zero Dataset rows and no Result event.
• Each populated row includes its source URL, canonical member URL, timestamp, status, page type, and Actor provenance.

This Actor is a public-page snapshot extractor, not an identity, employment, education, contact, or background-verification service. Source-selected content can be outdated, incomplete, approximate, or absent. Do not use a result as the sole basis for hiring, eligibility, credit, housing, insurance, or another consequential decision.

Run one profile lookup · Connect through the Apify API

Why Choose This API

A public LinkedIn member page can expose identity text, structured Person data, profile sections, article cards, and activity cards through different page surfaces. Their presence changes by profile and request. This Actor requests the public page, extracts the available elements, normalizes counts and URLs, and writes one consistently shaped person row only when a name can be parsed.

The contract is intentionally narrow. It accepts one member URL rather than a search query or URL list. A populated row always contains 19 top-level keys, but extracted scalar values can be null and array fields can be empty. That distinction matters: an empty experience array means no entries were parsed during the run, not that the person has never worked.

Provenance makes the snapshot reviewable. source_url records the accepted input, linkedin_url stores a canonicalized form without query parameters, processed_at identifies observation time, and processor identifies the Actor surface. Retain those fields when joining data to another system.

The Actor does not log in to a member account, use Sales Navigator or Recruiter data, search LinkedIn by name, reveal private contact information, confirm that two similar names represent the same person, or verify profile claims. It also does not promise current source access. Earlier and later runs for the same example profile can legitimately differ between one row and zero rows.

Quick Start Guide

Open the Actor input page and enter one HTTPS member-page URL. Company pages under /company/, job pages, search pages, and URLs on other domains are outside the public input contract.

{
  "linkedin_url": "https://www.linkedin.com/in/andrewyng/"
}

Start the run and wait for a terminal status. Then inspect the default Dataset rather than treating SUCCEEDED as proof that a profile was returned. In the verified success baseline, the Dataset contained one row and charged Actor Start 1 plus Result 1. Three later source-availability tests finished successfully with zero rows and Result 0.

For production adoption, store the run ID, Dataset ID, input URL, build number, status message, item count, charged-event counts, and timestamp together. A single successful profile does not demonstrate reliable coverage of other pages.

If the first run is empty, verify the URL shape and try the public source page later. Avoid immediately launching many duplicates, because each run can still incur Actor Start and metered usage even when no Result is written.

Input Parameters

Parameter	Type	Required	Supported public format	Behavior
linkedin_url	string	Yes	https://www.linkedin.com/in/{slug}/	One public LinkedIn member page. The schema rejects unsupported paths and malformed URL forms before execution.

Only this field is supported. There is no batch array, maximum-results setting, member-name search, company filter, keyword filter, date range, geographic filter, cookie field, credential field, or contact-enrichment switch. Trigger separate runs for separate member URLs.

The runtime trims and lowercases the submitted URL before requesting it. It validates the LinkedIn /in/ path and canonicalizes the successful row URL by removing query and fragment components. The public schema deliberately requires HTTPS and the standard www.linkedin.com member path, preventing ambiguous scheme-less inputs from entering the extraction path.

The input URL identifies the page to observe; it does not establish the subject's identity or authorize a particular use. Confirm your purpose, lawful basis, notice obligations, retention period, access controls, and platform permissions before processing personal data.

Output Data Schema

A populated Dataset row contains all 19 top-level keys in a fixed order. Extracted values remain source-dependent.

Group	Fields	Important boundary
Provenance	source_url, processor, processed_at, status, type	Describes this Actor observation, not external verification.
Identity	linkedin_url, name, headline, about, profile_image_url	Public source text and media; image links can expire.
Place and counts	location, follower_count, connection_count	Displayed snapshots; location is not a verified residence and counts can be capped or change.
Current-role signals	current_companies, job_titles	Derived from source structured data; arrays can be empty and are not employment verification.
History	experiences, education	Parsed visible entries, not guaranteed complete work or education histories.
Content samples	articles, activity	Source-visible samples, not complete publication or post archives.

This abridged sample uses values observed in the verified Andrew Ng baseline. Counts and arrays are snapshots, not fixed examples for future runs.

{
  "source_url": "https://www.linkedin.com/in/andrewyng",
  "processor": "https://apify.com/agentx/linkedin-profile-lookup?fpr=aiagentapi",
  "processed_at": "2026-07-23T11:50:20+00:00",
  "status": "success",
  "type": "person",
  "linkedin_url": "https://www.linkedin.com/in/andrewyng/",
  "name": "Andrew Ng",
  "headline": "DeepLearning.AI, AI Fund and AI Aspire",
  "location": "Palo Alto, California, United States",
  "follower_count": 2569909,
  "connection_count": 500,
  "current_companies": ["DeepLearning.AI", "AI Fund", "LandingAI", "Coursera", "Stanford University"],
  "job_titles": ["Founder", "Managing General Partner", "Executive Chairman"],
  "experiences": [{"title": "Founder", "company": "DeepLearning.AI"}],
  "education": [{"school": "Massachusetts Institute of Technology"}],
  "articles": [],
  "activity": []
}

Nested objects can include more keys than the shortened sample. Experience entries can include company and logo URLs, date text, parsed start/end years, and duration. Education entries can include school and logo URLs, degree, date text, and parsed years. Article and activity records can include URLs, source text, timestamps, media, and engagement counts. Any nested value can be absent where the source card does not expose it.

Integration Examples

The stable Actor ID is CUf5sseg9cEozOyhU, and the full name is agentx/linkedin-profile-lookup. With an Apify client, start the Actor using the input above, poll the terminal run status, and read items from defaultDatasetId. Treat zero items as a supported source outcome rather than a malformed API response.

For HTTP integrations, POST the input JSON to the Actor run endpoint using a server-side Apify token. Poll the returned run ID, then request Dataset items. Do not embed tokens in browser code, prompts, shared notebooks, URLs, or logs. A retry policy should distinguish transport failures from a successful zero-row run.

MCP clients should fetch current Actor details first and then call the Actor. A minimal call is:

{
  "actor": "agentx/linkedin-profile-lookup",
  "input": {
    "linkedin_url": "https://www.linkedin.com/in/andrewyng/"
  },
  "callOptions": {
    "memory": 128,
    "timeout": 300
  },
  "waitSecs": 45
}

If the wait period ends while the run is still active, poll that run rather than creating a duplicate. After completion, read the Dataset by ID and confirm its item count and key set. Persist raw snapshots or hashes only when your retention policy permits them.

Downstream consumers should test for null, empty arrays, and zero rows. Do not silently carry an old non-null value forward when a later run returns no row; represent source unavailability separately from a confirmed profile change.

Pricing & Cost Calculator

The Actor uses pay-per-event pricing. The Apify Console and the charged-event record for your run are authoritative.

Event	Trigger	Free	Bronze	Silver	Gold	Platinum	Diamond
Actor Usage	Runtime, proxy, and storage consumption	$0.00001 per usage unit	same	same	same	same	same
Actor Start	Actor starts; platform memory rules affect counted starts	$0.01	same	same	same	same	same
Result	One person row written to the default Dataset	$0.15000	$0.14500	$0.14000	$0.13500	$0.13500	$0.13500

At the Free-tier Result rate, one populated run contributes $0.15000 for Result plus Actor Start and metered Actor Usage. A successful zero-row run has no Result event but can still incur Start and Usage. The observed baselines matched this boundary: the populated run charged Start/Result 1/1, while later zero-row runs charged 1/0.

Estimate spend from attempted runs as well as expected populated rows. If 100 attempts all return a row, Result charges at the Free tier contribute 100 × $0.15000 = $15.00, before Start and Usage. If source availability yields fewer rows, Result charges fall, but every started attempt can retain its other event costs.

Set any maximum-total-charge value high enough for Start, Usage, and the expected Result. A billing cap is not a source-availability control and does not make missing profiles appear.

Use Cases & Applications

Recruiting teams can collect public profile snapshots for candidates already identified through an authorized workflow. Keep provenance and require human review. Parsed experiences and education are research leads, not verified credentials, and should not replace consent, references, or formal checks.

Account researchers can use current-company and title signals to prepare a source-linked brief. The arrays can be empty or stale, and a displayed person should not automatically be treated as a buying authority or a permissible outreach target. Apply suppression, opt-out, and jurisdiction-specific outreach rules.

Market researchers can observe public career and content signals across a defined, lawful panel. Follower and connection counts are snapshots, while articles and activity are samples. Use consistent observation intervals and do not interpret a missing record as a confirmed real-world event.

AI systems can retrieve a current source snapshot before answering a question about a public profile. The model should cite the observation time, separate source claims from verified facts, and avoid inferring sensitive traits from names, photos, education, location, or activity.

Data-quality teams can compare key presence, array lengths, and source text between runs. Changes should enter a review queue because page variants, transient source access, parsing changes, and media expiry can resemble real profile changes.

FAQ

Does a successful run always return a profile?

No. A run can finish successfully with zero Dataset rows when the page cannot be fetched or a name cannot be parsed. Check both terminal status and item count.

Are all 19 fields guaranteed to contain values?

A populated row contains all 19 top-level keys, but extracted scalar fields can be null and array fields can be empty. Only the row envelope is fixed.

Is the experience or education history complete and verified?

No. The Actor parses visible source entries. Visibility, markup, and page content can be incomplete, and the Actor does not verify claims with employers or schools.

Is connection_count exact?

Not necessarily. LinkedIn can display capped or formatted values. Treat both connection and follower counts as observation-time signals.

Does this return emails, phone numbers, or private account data?

No supported output field represents an email or phone number. The Actor reads a public member-page surface and does not accept account credentials.

Can I submit many URLs in one run?

No. Submit one linkedin_url per run and preserve separate evidence for each outcome.

Trust & Certifications

The readiness baseline used the current cloud Build with 128 MB memory. One earlier Andrew Ng run returned a 19-key Dataset row, while three later tests against public member URLs returned zero rows without Result events. This evidence supports the documented populated-row and zero-row branches; it does not establish future success rate.

No accuracy, completeness, freshness, identity, employment, education, contact, privacy, legal-compliance, uptime, or third-party certification is claimed. LinkedIn controls the source, Apify controls the cloud platform, and this Actor normalizes what it can observe. The Actor is not an official LinkedIn integration.

Production users should monitor run status, Dataset count, key presence, null and empty-array rates, array cardinalities, count changes, charged events, runtime, and unexpected source text. Use access controls and retention limits appropriate for personal information. Protect tokens and avoid placing unnecessary profile content in support messages.

Legal & Compliance

Use this Actor only for purposes, data, and volumes permitted by applicable law, contracts, and platform rules. You are responsible for legal basis, transparency, data-subject rights, retention, security, transfer safeguards, outreach restrictions, and decisions made from the output. Public visibility does not remove privacy, employment, intellectual-property, or database-right obligations.

Review LinkedIn's User Agreement and Privacy Policy, plus the Apify General Terms and Conditions, Actor Terms, and Privacy Policy. These links are informational and are not legal advice.

Do not use the output to infer sensitive traits or to make solely automated employment, credit, housing, insurance, immigration, education, or eligibility decisions. Establish a documented human-review and correction process where personal data affects someone.

Related Tools

For adjacent LinkedIn company context, use LinkedIn Company Lookup. For broader vacancies, use All Jobs Scraper. For a source-dependent employer-page snapshot from another jobs platform, use JobStreet Company Profile.

Enrich an organization already identified in the profile with Google Maps Store Scraper, compare public topic interest with Google Search Trends API, or inspect a public X account with Twitter Info Scraper. Each Actor has independent inputs, output semantics, pricing, and source risks.

Support & Community

For a reproducible issue, send the run ID, redacted or already-public member URL, build number, terminal status message, Dataset item count, top-level key set, and charged-event counts. State whether the issue concerns input validation, source availability, a null value, a nested array, count parsing, event charging, or API integration.

Never send an Apify token, proxy password, database credential, private Dataset, cookie, account credential, or unrelated personal data. Remove unnecessary personal information from screenshots and logs.

Use AgentX support on Telegram for Actor-specific help and the Apify community for platform questions.

Last updated: July 23, 2026.