KDA Blog
Notes from the
kernel design loop
Results, failure modes, and lessons from building agents that write, verify, and tune GPU kernels.
LATEST POST
KDA²: Kernel Design Agents (KDA) optimize Kimi Delta Attention (KDA)
Our agents wrote Kimi Delta Attention kernels that run up to 2.96× faster than FlashKDA on B300 with a tenth of its state error. Here is how, and how the agents tried to cheat along the way.
Read the postALL POSTS
1 post