Meetup In 3 days
November
Friday
13:00–17:30 2026 · Singapore time
You sign up with the host. foundercal only links to it.
Your event? Manage this listing Report a change
vLLM Meetup Singapore: Where the World’s Leading Open-Source AI Inference Engine Meets Asia’s Builders
vLLM is the open-source engine behind much of today’s large-scale AI serving. It was born at UC Berkeley, is now hosted under the PyTorch Foundation, and is used from frontier labs to hyperscalers to fast-growing startups. It supports hundreds of model architectures and runs on a wide range of accelerators.
We are bringing the vLLM community to Singapore for a half-day technical meetup!
What attendees will experience:At this meetup, engineers from Inferact (founded by vLLM’s creators and core maintainers) and NVIDIA will share what they’re building right now, from high-throughput prefill and RL training to running vLLM on workstation-class Blackwell GPUs and NVIDIA’s next-generation Vera Rubin platform. They’ll be joined by Alibaba Cloud, who will bring a full-stack view spanning models, training, inference, and model-as-a-service. The event is hosted at SQ Collective, a Singapore co-working space for AI founders and builders. If you build, deploy, or research LLM systems, this is a rare chance to hear from the people shaping the engine and to talk with them in person.Agenda:13:00 ~ 13:30 Registration13:30 ~ 13:40 Opening Speech 13:40 ~ 14:10 Optimize Pipeline Parallelism for High throughput Prefill in vLLM - Thien Tran, Member of Technical Staff, Inferact14:10 ~ 14:40 vLLM on SM12x: From Usable to Efficient - Jie Fang, DevTech Compute Engineer, NVIDIA14:40 ~ 15:10 Enterprise Scale Agent Infra - Sean Zhan, Senior Director Alibaba Cloud International 15:10 ~ 15:30 Break15:30 ~ 16:00 Faster Effective RL Training: High‑Performance, Bit‑Exact Training‑Inference Consistency, and Agent‑Driven Engineering - Ao Shen, Member of Technical Staff, Inferact16:00 ~ 16:30 vLLM NVIDIA Vera Rubin Update - Hao Lu, Software Engineer, NVIDIA16:30 ～ 17:00 Making World Models Interactive: Real-Time Inference with vLLM-Omni - Taichang Zhou, vLLM-Omni Committer17:00 ~ 17:30 Networking📍 Venue: SGInnovate 32 Carpenter Street Singapore 059911📅 Date: 13 Nov 2026, 13:00 - 17:30 SGT 🎟 Free to attend. Registration required; approval is subject to venue capacity.
About SQ Collective
SQ Collective is a co-working space in Singapore for founders and builders working with AI. Fridays are open co-working days, with a short shared session called The Stage at 1.30pm. People come to work, learn something practical, and go back to building.
Proudly supported by Singapore Global Network
Supported by Singapore Global Network (SGN), a division under Singapore Economic Development Board (EDB). Join SGN to become a member of our global network of professionals and friends with Singapore at its heart. Learn more at www.singaporeglobalnetwork.gov.sg.
Full text on Luma .
Founders pitch
Investors in the room
Investors in the room
Investors in the room
