[mendeza]

For anomaly detection, does your team have to worry about concept drift and data distribution shifts? If so, how do you combat that?

[michaelgygli]

Great question. Yes, that's a big topic for us. We closely monitor production metrics to detect things like input changes and output drift. This allows us to automatically detect if something changes, so that we can trigger retraining and redeployment.

[gyglim]

We built a lot of this in-house though, as we didn't find suitable existing tooling for that