SGLang Summit Workshop by Modal: Inference from First Principles
Hosted by SGLang Summit 2026 Community Events + LMSYS Org + Modal Events + Will Reese
View organizer listing
What to look forward to
Modal Member of Technical Staff Gilford Ting will use first-principles models to explain why modern LLM architectures and serving systems are designed for efficient inference, and what those choices suggest for the next generation. The 60-minute session is aimed at attendees with moderate LLM inference experience; a SGLang Summit Full or Expo pass is required.
Inside the event
A SGLang Summit ticket is required to attend the event and get approved on Luma. Both Full and Expo passes are accepted.
Ticket 🎫 https://luma.com/sglang-summit
The room seats 45, and seating is first come, first served, so registering doesn't guarantee a seat. Please arrive early to get a spot.
Modern LLMs are increasingly co-designed with the systems that serve them, with architectures built from the ground up to be efficient for inference. But the reasoning behind these design choices is often reduced to graphs and benchmarks showing numbers going up, with little explanation of why they work. In this talk, we'll start from first principles; using simple analytical models, we'll derive key properties of today's models and serving systems, and explore what they suggest about the next generation of inference.
Session Details
- Duration: 60 minutes
- Level: Moderate experience in LLM inference serving. Familiarity with concepts like sharding and hybrid attention will be helpful.
- What to bring: No laptop needed
Speaker
- Gilford Ting, Member of Technical Staff, Modal
Fort Mason
Fort Mason, San Francisco, CA 94109, USA. Confirm the venue with the organizer before traveling.
Open in Google Maps ↗Details come from the event listing and the organizer page; registration is managed by the organizer.
Last checked: Nov 13, 2026. Confirm current availability before making plans.
Organizer page checked: Nov 13, 2026. Check registration for the latest details.