Byron Hsu

At a glance. Byron Hsu is listed as Merge oncall for prefill-decode (PD) disaggregation at SGLang project, with 1 other current role listed. The profile sits among the builders and open-source contributors. Fields listed: Training systems and Inference and serving. The source lists 1 official page and open-role pages for xAI.

Summary

Led Liger Kernel, open-source Triton kernels for memory-efficient LLM training and fine-tuning.

Known for: Liger Kernel; SGLang.

Works at

• SGLang project 2 people in the wiki
• Periodic Labs 2 people in the wiki

Other documented roles

Not confirmed as current in the source profile.

• Liger Kernel project lead, LinkedIn
• Previous role listed on GitHub profile, xAI

Open roles

• Open roles at xAI on USASI ↗

Related people

• Ying ShengAlso at SGLang project
• Hao ZhangInference and serving · Training systems
• Peter SalankiTraining systems · Inference and serving
• Zhuohan LiInference and serving · Training systems
• Robert ShawInference and serving · Training systems
• Vipul Ved PrakashInference and serving · Training systems
• Kunle OlukotunTraining systems · Inference and serving
• Daniel HanTraining systems · Inference and serving
• Michael IntratorTraining systems · Inference and serving
• Furu WeiInference and serving · Training systems

Source and corrections

Read the full profile on USASI ↗

Facts on this page come from that USASI profile (reviewed Oct 21, 2026); data last checked Oct 23, 2026. See something wrong? Suggest a correction.