Hi NVIDIA team,

I’m looking to get routed to the right engineering owner for a short 20–30 min technical screen.

Public-safe evidence from our H100 runs:

• Explicit (N\times N) fp16 materialization becomes infeasible at large N (measured CUDA OOM at N=500,000; attempted allocation is hundreds of GiB).
• An indexed O(N) retrieval path continues to operate at the same N (no (N^2) matrix construction).
• Memoization on repeated queries yields a large hot-path speedup (example: 863×).

We can share a public-safe PDF first; under NDA we can provide a reproducible runbook + evidence bundle (logs/scripts) and do a controlled review (no public source release).

Who is the right maintainer/engineer to speak with, or what is the best way to contact the correct CUDA/perf/inference integration owner?

Thanks,
Stanislav Byriukov