Ulysses, a novel sequence parallelism technique, boosts long-context LLM inference performance with 3.4x lower latency and better GPU efficiency.
APR 03, 2025|14 min read
Mert Hidayetoglu
Mert Hidayetoglu is a software engineer at Snowflake’s AI systems research. Previously, he was a postdoctoral scholar at the CS department of Stanford University and worked on programming models for exascale systems. Mert obtained his PhD in electrical and computer engineering from University of Illinois in 2022, where his research was on optimization of sparse memory accesses and communications for solving large-scale inverse problems. He was a Givens Fellow at Argonne National Laboratory in 2018, the leading author of SC20 best paper, and recipient of ACM/IEEE-CS George Michael Memorial HPC Fellowship in 2021.