Vectris Labs was honored to be invited to the Supermicro + AMD Enterprise AI & HPC Summit in New York City.
The gathering brought together CAIOs and leaders from organizations including UBS, Wallaroo AI, Aible, AMD, Supermicro, Scale AI, and Squarepoint Capital to discuss what is actually working – and what is not – in enterprise AI deployments.
The operating constraints are becoming clearer.
The conversation repeatedly returned to inference efficiency, GPU utilization, cost per token, power density, production deployment, and the challenge of tying infrastructure spend to business outcomes.
Organizations are investing heavily in GPU infrastructure while still confronting low utilization, visibility gaps, and returns that do not match expectations.
Inference governance belongs in the stack.
The discussions reinforced our view that inference governance is not a nice-to-have. It is the missing layer between optimized serving software and the economic outcomes infrastructure teams are expected to deliver.
Thank you to the Supermicro and AMD teams for organizing the event and to IgniteGTM for bringing together the right operators and builders.
Adapted from Vectris Labs LinkedIn updates about the Enterprise AI & HPC Summit NYC.