Bot defense that doesn't punish your SOC team

The implementation was generally smooth, but it required a few upfront realities to be managed carefully. One of them is that baseline tuning was not optional. Out of the box, the behavioral models were solid but needed a tangible amount of fine tuning. During staging, we underestimated how much of our staging traffic was skewed by internal users and bots. as a result, the initial training data we gave F5 Distributed Cloud Bot Defense didn't represent real world traffic patterns.

I highly rate them because of their responsiveness and technical competency. Every time we've submitted a ticket, we've had access to deep technical resources within a day. They don't just quote documentation, they actually troubleshoot with us. That's rare

I'd strongly recommend it, but with a few caveats depending on how mature the team is with behavioral based security tools. One of our fintech clients was getting hit with low volume, widely spread login attempts, below our rate limiting thresholds. F5 Distributed Cloud Bot Defense was able to flag abnormal input timings, inconsistent device fingerprinting and high entropy in field population behavior. You can only imagine the wave of downstream account lockouts this saved the client.
On the other end we had a client with a real time trading platform using Graphql over websockets. F5 Distributed Cloud Bot Defense wasn't able to tap into that stream natively. we had to reverse engineer a proxy layer to inspect events. It worked but it was clunky and not officially supported