what's it like working with robot arms with that kind of latency? Any AI stuff (ML models etc) on board?

The RPO and docking are autonomous specifically because it's hard to deal with the latency.

Note that the algorithm stack for almost any spacecraft cannot rely on modern compute. Single-core sub-1 GHz processors are the standard. Very few GPUs have ever been in space, in any capacity, and none that I know of in a mission-critical role.