AnthropicOther·60 minMembers
Performance Modeling — Matmul Arithmetic Intensity
Members only
An onsite round for the Performance Engineer track: estimate wall-clock time for matrix multiplications of varying `(m, k, n)` on a given A100 40 GB GPU, then extend to a 72-layer pipeline alternating...
Perf Eng
MLE
Infra Eng
gpu
a100
matmul
arithmetic-intensity
performance-modeling
ml-knowledge
hard
Frequency
Low
Last asked
2026-09-17
Stage
onsite-research · tech-screen
Log in to continue reading the full content
Comments
Sign in to join the discussion
Loading...
