Hyperbach

Data feeds you can build on: jobs, hiring clients and ads, delivered as clean, typed rows. We build for scheduled pipelines, internal tools and AI agents as much as for people.

Upwork Job Scraper. New Upwork postings with client quality signals and 60 filters. A cursor-based notification mode delivers each job once and never skips one.

Job board history. Himalayas, Ashby, Greenhouse, RemoteOK, We Work Remotely and Naukri, plus a combined feed of 7 boards. Each keeps the full board and each posting's expiry history, not only today's listings.

Ad intelligence. Google Ads Transparency with ad copy and image OCR, and the Meta Ad Library with EU reach.

Company and book data. Gregslist companies (SaaS founders and their websites), and Goodreads books, ratings and reviews.

Tested against the live build. More than 1,300 automated tests, and 82 acceptance scenarios that run the published Actor on real inputs before and after each release.

Every run is measured. Status, rows, timing and upstream health are recorded for every run, and failures page an engineer. Servers, services and response times are checked every 5 minutes, around the clock, with alerts.

Honest failures. A failed run fails visibly and charges nothing beyond Apify's start fee. A partial run says so in its status line. Inputs are read leniently, and a real input error says exactly what to change.

Shipped daily, often several times a day. About 300 changes last week across the fleet; the Upwork scraper went from build 1.1.63 to 1.1.76 in three days. A buyer's failed run is usually fixed within the hour.

Questions, feature requests, or a data source you need: apify@hyperbach.com