Mercor Scraper

Extract live contractor and expert gig listings from Mercor, the AI-recruiting/talent marketplace that matches subject-matter experts, freelancers and knowledge workers to paid work for AI labs and their client companies (data labeling, model evaluation, occupation studies, expert interviews, and part-time or hourly contract roles). Get every live listing as clean structured JSON: title, hiring company (when disclosed), pay rate, category, location, commitment type and posted date.

No login. No cookies. No browser automation. The actor talks directly to Mercor's own public JSON API — the same endpoint the work.mercor.com website calls — so it is fast, cheap and stable.

What this actor does

Mercor runs a live marketplace of paid gigs: AI training data annotation, expert evaluation of model outputs, one-off occupation surveys, and ongoing part-time/full-time contract roles across dozens of professional categories (law, medicine, finance, software engineering, data analysis, and more). This actor pulls the full current set of active listings in one call and lets you filter it down by keyword, category, commitment type, location and minimum pay rate — then returns one row per listing with:

• Listing title and full description
• Hiring company name (Mercor blinds the company on roughly 6 in 10 listings until a candidate applies — those come back as null, which is expected, not a scraping gap)
• Category ("listing domain" — e.g. Software Engineering, Law, Medicine, Data Analysis)
• Location and work arrangement (remote/onsite)
• Commitment type (hourly, part-time, full-time) and expected hours/week
• Pay rate range and frequency
• Whether the listing offers equity
• Eligible countries/regions, if the listing is geographically restricted
• Posted date and a direct, canonical listing URL

Why use it

• AI-labor market research — track how many paid AI-training/expert gigs are live at any moment, and in which categories and price bands.
• Rate benchmarking — Mercor publishes hourly rates directly; use this to benchmark what AI labs are paying subject-matter experts by domain (law, medicine, finance, engineering).
• Recruiting and sourcing intel — see which client companies are hiring through Mercor and for what kind of work.
• Lead generation for freelancers/agencies — a live feed of open remote gigs filtered by category or minimum rate.
• Competitive intelligence — Mercor is one of the fastest-growing AI-recruiting marketplaces; this actor gives you a repeatable snapshot of its live listing volume and mix.

How to scrape Mercor data

1. Open the actor and set a Search Keyword (for example data engineer, legal, medicine or leave the default). It matches against both the listing title and full description.
2. Optionally pick a Category (Listing Domain) such as Software Engineering, Law, Medicine or Data Analysis, or leave it on "All categories".
3. Optionally filter by Commitment Type (hourly / part-time / full-time), Location (e.g. Remote, United States) and Minimum Pay Rate.
4. Set Max Results to how many listings you want back.
5. Click Start.

The actor makes a single call to Mercor's listings-explore-page API, which returns every currently active listing (246 at the time this actor was built), then applies your filters client-side — Mercor's own API doesn't support server-side search or pagination parameters, so filtering happens in the actor for reliable, complete results. Export as JSON, CSV or Excel.

Input

Field	Type	Description
searchQuery	string	Keyword matched against title + description (case-insensitive). Default: data.
listingDomain	string	Filter by category (e.g. Software Engineering, Law, Medicine). Empty = all categories.
commitment	string	Filter by hourly, part-time or full-time. Empty = all.
locationFilter	string	Substring match on the location field (e.g. Remote, United States). Empty = all.
minRate	integer	Minimum pay rate (max of the listing's rate range must be ≥ this number). Default: 0 (no filter).
maxResults	integer	Maximum number of listings to return. Default: 50.
proxyConfiguration	object	Apify proxy settings. Mercor's API has no anti-bot, so any setting (including none) works.

Example input

{
    "searchQuery": "data engineer",
    "listingDomain": "Data Analysis",
    "commitment": "",
    "locationFilter": "",
    "minRate": 0,
    "maxResults": 50
}

Output

Each dataset item looks like this (real output from a live test run):

{
    "listingId": "list_AAABn4eLz1zT1QqB0LtGSr92",
    "jobTitle": "Law Expert",
    "company": "Deeptune",
    "listingDomain": "Law",
    "location": "Remote",
    "workArrangement": "remote",
    "commitment": "hourly",
    "hoursPerWeek": 40,
    "salaryMin": 75,
    "salaryMax": 150,
    "salaryFrequency": "hourly",
    "salaryCurrency": "USD",
    "offersEquity": false,
    "eligibleLocations": ["USA"],
    "postedDate": "2026-07-22T01:58:35",
    "description": "## About the Role\n\nMercor is partnering with a leading AI lab to train frontier models on high-quality legal reasoning data...",
    "url": "https://work.mercor.com/jobs/list_AAABn4eLz1zT1QqB0LtGSr92",
    "scrapedAt": "2026-07-22T15:05:21.962Z"
}

Output fields

Field	Type	Description
listingId	string	Mercor's internal listing ID
jobTitle	string	Listing title
company	string | null	Hiring company name, or null on blind listings
listingDomain	string	Category (e.g. Software Engineering, Law, Medicine)
location	string	Location as listed (usually "Remote")
workArrangement	string | null	remote / onsite / hybrid
commitment	string	hourly / part-time / full-time
hoursPerWeek	number | null	Expected weekly hours, if specified
salaryMin	number | null	Minimum pay rate
salaryMax	number | null	Maximum pay rate
salaryFrequency	string | null	Pay rate frequency (usually "hourly")
salaryCurrency	string	Rate currency (USD unless otherwise stated)
offersEquity	boolean	Whether the listing offers equity
eligibleLocations	array	Country/region codes eligible to apply, if restricted
postedDate	string	ISO date the listing was posted
description	string	Full listing description (markdown source)
url	string	Canonical listing URL on work.mercor.com
scrapedAt	string	ISO 8601 timestamp of collection

Cost estimate

This actor makes exactly one HTTP call per run (Mercor returns its entire live listing set in a single response), so runs are extremely cheap and fast — a full run over hundreds of listings typically finishes in a few seconds of actual compute. Exact cost depends on your Apify plan and the number of results you request.

Limitations

• Company names are blind on many listings. Mercor hides the hiring company's identity on a large share of listings (roughly 6 in 10 at the time of writing) until a candidate applies — this is a Mercor product decision, not a scraping gap, and shows up as company: null.
• No server-side filtering on Mercor's end. Mercor's API always returns the full active listing set and ignores query parameters; this actor filters client-side, which is reliable but means the API call itself always fetches everything before your filters are applied.
• Only currently active listings. Mercor's API only exposes live listings, not historical or closed ones — run the actor on a schedule if you need a time series.
• Rates are as published by Mercor, not verified independently; a small number of listings omit a rate entirely (salaryMin/salaryMax: null).

Related scrapers

Building a broader jobs dataset? These sibling actors follow the same clean-JSON, no-login approach:

• NoFluffJobs Scraper — European tech job listings with transparent salary ranges
• Greenhouse Jobs Scraper — job postings from any company's public Greenhouse career board
• Xing Jobs Scraper — DACH-region job listings (Germany, Austria, Switzerland)
• Totaljobs Scraper — UK job listings
• Jobs.ch Scraper — Swiss job listings in German, French and English