We used to beat bot protection.
Now we build it.
Centinel is a small team of reverse engineers and bot developers based in Berlin. Before we switched sides, we spent years building the crawlers that site owners were trying to stop. That's the whole reason we exist: we got tired of watching detection fail against tools like the ones we used to build, and decided to build protection informed by that experience.
We approach detection from both sides: how automation gets through, and what a website can observe and control.
Featured by news outlets and industry bodies
Detection has to move faster than the attacker
Browser tools, network paths, and evasion methods change. We investigate those changes, review detection evidence, and test how controls affect both automation and legitimate visitors.
Below the JavaScript layer
Connection evidence adds context to browser signals. Its usefulness depends on which connection the integration observes and what a proxy or CDN changes.
Test against real tools
Our published benchmark covers eight scraping and browser-automation tools. Use its recorded outcomes to choose what to test against your own workflows.
Dynamic by design
Browser checks and evasion techniques evolve. Detection needs repeatable tests and ongoing review as those tools change.
Measure the outcome
Review protected resources, completed user workflows, and false positives together. A higher block count alone does not show that a policy works.
Who writes our technical guides?
Frederick Jahn
Managing director
Frederick writes Centinel's technical guides on crawler identity, browser automation, and bot-protection evaluation. His articles connect operator documentation with practical investigation and testing methods.
- Browser-automation detection guide
- Crawler-identity verification guide
- Bot-protection evaluation scorecard
For questions about a guide or to report a correction, contact the team with the article URL and supporting evidence.
Come build bot protection for site owners
We're a small team working on a hard problem: catching the crawlers that look human. If that sounds like your kind of problem, we'd like to hear from you.
See open roles