Zyte, formerly Scrapinghub, is a web-scraping company founded in 2010 by Pablo Hoffman and Shane Evans that also maintains Scrapy, one of the most widely used open-source scraping frameworks.
What it is
Zyte API is the commercial product: an all-in-one endpoint that picks the right proxy for a target site's complexity and rotates it automatically, so a customer doesn't manage a proxy list by hand. It grew out of what used to be a separate product, Crawlera, now folded in as the API's proxy mode and still available to Scrapy users directly through the scrapy-zyte-smartproxy middleware. More recently, Zyte added spider templates and AI-assisted extraction through Scrapy Cloud, aimed at pulling structured data without hand-writing selectors for every site.
How it gets used to scrape
Zyte sits underneath a lot of production scraping that predates the current wave of AI agents: price monitoring, catalog extraction, and general data collection run through Scrapy, with Zyte's proxy layer handling the part where a target site starts blocking a single IP. The AI-assisted spider templates lower the bar further, letting a customer point Zyte at a site without writing extraction logic themselves.
Why it's hard to catch with old checks
Zyte's whole value proposition is proxy rotation tuned to the target site, which is specifically designed to defeat IP-based rate limiting and reputation blocking. A detector that keys mainly on a single IP's request volume is exactly what this product is built to get past, since the request volume from any one proxy in the rotation stays low. What doesn't rotate away is the shape of the requests themselves and how consistently they come from automation built on Scrapy's request pipeline rather than a browser.
Centinel detects it
In our September 2026 benchmark, Centinel identified every session run through Zyte's infrastructure. Because Centinel scores request and behavioral signals across a session rather than relying on IP reputation alone, proxy rotation doesn't remove the tell.
Key takeaways
- Zyte maintains Scrapy and sells Zyte API, an all-in-one scraping product built around automatic proxy selection and rotation.
- It underpins a lot of pre-AI-agent scraping infrastructure: price monitoring, catalog extraction, structured data collection.
- Proxy rotation defeats IP-based blocking by design; it doesn't change the request-level and behavioral signals Centinel scores instead.