doing
gpu architecture at
apple
places
interned on gpu architecture team at
apple
interests
deep learning
computing paradigms + architectures
co-optimising software + hardware
projects
mSight — a terminal-based performance monitor for apple silicon [
1]
tinyflash — a minimal implementation of flash-attention [
2]
tinyoptimizer — a minimal implementation of a superoptimizer for tensor programs [
3]
claude-safely-skip-permissions — run claude autonomously without dangerous commands [
4]