Upwork Job Finder

Overview

Upwork Job Finder is an Apify Actor that scrapes freelance job listings from Upwork's public job search for any keyword you choose. It bypasses Cloudflare's anti-bot challenge using a stealth headless browser, extracts the embedded Nuxt application state, and returns clean, structured job data ready for analysis, alerting, or integration into downstream workflows.

Typical use cases:

• Build a daily feed of new freelance opportunities matching your skills.
• Power a job-alert email or Slack notification system.
• Track market rates, demand, and required skills for a niche over time.
• Feed structured job data into a CRM, Airtable base, or BI dashboard.

Features

• Search any Upwork query (skills, keywords, or phrases).
• Paginate through results one page at a time (available to all users).
• Multi-page range scraping (paid users only): fetch up to 10 pages in a single run using pageStart and pageEnd. Free-tier users are limited to single-page scraping via the page field.
• Automatic Cloudflare Turnstile solving via a stealthy headless browser.
• Optional Apify Proxy support via proxyConfiguration.
• Returns full job metadata: title, description, budget, skills, client info, posting timestamps, and more.
• Output is written to the Actor's default dataset with a typed schema and a pre-built table view.

Input

The Actor accepts either a plain search query or a full Upwork searchUrl — not both. Use query for a quick keyword search, or build a URL on Upwork's search page (filters for budget, experience level, duration, client history, etc.) and paste it into searchUrl. Any Upwork URL filter works automatically.

Field	Type	Required	Default	Description
query	string	One of query/searchUrl	—	Plain keyword search (3–100 chars), e.g. "Airtable Automation".
searchUrl	string	One of query/searchUrl	—	Full Upwork search URL. Must match https://www.upwork.com/nx/search/jobs/?.... Any other URL is rejected.
page	integer	No	1	Result page to fetch when pageStart/pageEnd are not provided.
pageStart (paid only)	integer	No	—	First page in the range to fetch. Must be ≥ 1. Available to paying Apify users only. The run fails immediately if a free-tier user provides this field.
pageEnd (paid only)	integer	No	—	Last page in the range to fetch. Must be ≥ pageStart. Available to paying Apify users only. The run fails immediately if a free-tier user provides this field.
proxyConfiguration	object	No	—	Optional Apify Proxy configuration. Use the proxy editor in the Apify UI.

Validation: the run fails fast if both query and searchUrl are set, if neither is set, or if searchUrl is not a valid Upwork search URL. The page range is capped to 10 pages per run.

Free vs. paid plans. Range-based scraping (pageStart/pageEnd) is restricted to paying Apify users. Free-tier users can still scrape jobs — just one page per run via the page field. If a free user passes pageStart or pageEnd the run is rejected with a clear error message. Upgrade your Apify plan to unlock multi-page runs.

Example: simple keyword query

{
  "query": "Airtable Automation",
  "page": 1
}

Example: full search URL with filters

{
  "searchUrl": "https://www.upwork.com/nx/search/jobs/?client_hires=0&contractor_tier=1&duration_v3=semester&nbs=1&q=ai%20automation&sort=relevance%2Bdesc",
  "page": 1
}

Example: with Apify Proxy

{
  "query": "AI engineer",
  "page": 2,
  "proxyConfiguration": {
    "useApifyProxy": true,
    "apifyProxyGroups": ["RESIDENTIAL"]
  }
}

Example: page range (capped to 10 pages, paid users only)

{
  "query": "Airtable Automation",
  "pageStart": 1,
  "pageEnd": 5
}

Note: this input only works on a paid Apify plan. Free-tier users should use the single-page page field instead, and loop over runs if multiple pages are needed (see Paginating through all results).

Output

Each run pushes one record per job to the default dataset. A full JSON Schema is defined in .actor/dataset_schema.json, and an Overview table view is available in the Apify console for quick scanning.

Key output fields

Field	Type	Description
uid	string	Unique job identifier.
ciphertext	string	URL slug. Build the job URL with https://www.upwork.com/jobs/{ciphertext}.
title	string	Job title (may include <span class="highlight"> markup around matched terms).
description	string	Full job description (may include highlight markup).
type	integer	1 = Fixed-price, 2 = Hourly.
durationLabel	string	Expected project duration, e.g. "1 to 3 months".
engagement	string	Engagement key, e.g. usnuxt_Engagement_421.partTime.
tierText	string	Required experience tier (entry / intermediate / expert).
amount.amount	number	Fixed-price budget in USD (0 for hourly jobs).
hourlyBudget	object	{ min, max } hourly rate in USD (both 0 for fixed-price jobs).
publishedOn	string (ISO 8601)	When the job was published.
attrs	array	Skill tags attached to the job, each with prefLabel and a highlighted flag.
client	object	Client info: country, payment verification, total spent, reviews, and feedback rating.

Sample output (truncated)

{
  "uid": "2054172332114949360",
  "ciphertext": "~022054172332114949360",
  "title": "Senior AI/ML Engineer for LLM and RAG Systems",
  "description": "We are seeking a highly skilled Senior AI/ML Engineer ...",
  "publishedOn": "2026-05-12T12:11:45.608Z",
  "type": 2,
  "durationLabel": "More than 6 months",
  "engagement": "usnuxt_Engagement_421.fullTime",
  "tierText": "jsn_Expert_207",
  "hourlyBudget": { "min": 30, "max": 60 },
  "amount": { "amount": 0 },
  "attrs": [
    { "prefLabel": "Machine Learning", "highlighted": true },
    { "prefLabel": "Artificial Intelligence", "highlighted": true },
    { "prefLabel": "Python", "highlighted": false }
  ],
  "client": {
    "location": { "country": null },
    "isPaymentVerified": false,
    "totalSpent": null
  }
}

Usage

Run on the Apify platform

1. Open the Actor in the Apify console.
2. Set your query and page values in the input form.
3. Click Start. Jobs appear in the Dataset tab when the run finishes.
4. Export the dataset as JSON, CSV, Excel, or HTML, or fetch it via the Apify API.

Run via the Apify API

curl -X POST "https://api.apify.com/v2/acts/<ACTOR_ID>/runs?token=<APIFY_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{ "query": "Airtable Automation", "page": 1 }'

Run locally

Requires Python 3.11+ and the Apify CLI.

$apify run

Edit storage/key_value_stores/default/INPUT.json to change the input.

Paginating through all results

Free users: the Actor fetches one page per run via the page field. To collect multiple pages, schedule or loop the Actor while incrementing page until an empty dataset is returned:

for page in 1 2 3 4 5; do
  apify call <ACTOR_ID> --input "{\"query\":\"Airtable Automation\",\"page\":$page}"
done

Paid users: use pageStart/pageEnd to fetch up to 10 pages in a single run, which is significantly faster than looping (the Cloudflare challenge is solved once and reused across all pages):

$apify call <ACTOR_ID> --input "{\"query\":\"Airtable Automation\",\"pageStart\":1,\"pageEnd\":5}"

For ranges larger than 10 pages, paid users can still loop runs, incrementing pageStart/pageEnd by 10 each iteration.

Limitations

• Only public search results are returned — no login-gated data.
• Free-tier users are limited to single-page scraping (page only). The pageStart/pageEnd range inputs require a paid Apify plan and will cause the run to fail on free accounts.
• Some client fields (totalSpent, country, isPaymentVerified, review counts) may be null in the search-listing payload. Fetch the individual job page for the full client profile.
• title and description contain Upwork's search-highlight HTML markup; strip it client-side if you need plain text.
• Heavy use of one IP may trigger Cloudflare. Use Apify Proxy for high-volume scraping.

Support & feedback

Found a bug or have a feature request? Open an issue on the Actor's source repository or contact the maintainer through the Apify console.

Wanna hire me? I'm a freelance software developer specializing in web scraping, automation, data engineering, web development, and much more. Check out my Upwork profile