Interviews + Prep

Spent most of this week prepping for the two technical interviews for the Princeton RSE job.

Hack Session

Started on Issue #232 which is partially addressed in PR #359. Still some bugs in it, the PPMC kernel is generating some weird results. Early performance gains are ~50-70% on a V100 which is great, hopefully we’ll have even better improvements on the MI250X.

This post is licensed under CC BY 4.0 by the author.