Real-time AI at the Edge: Inference, Kernels and Accelerators
Tokyo AI evening of three talks on low-latency edge AI: GPU profiling, portable inference runtimes and Jetson deployment.
- When
- Fri, November 13, 2026 · 18:00–21:00 JST
- Where
- Tokyo, Japan · In person
- Region
- Kanto (Tokyo)
- Organizer
- Tokyo AI
- Language
- EN
- Source
- Luma
Summary
Tokyo AI (TAI) hosts a technical evening on making AI inference fast when compute, memory and power are limited. Three 30-minute talks with Q&A cover GPU profiling and inference bottlenecks, portable inference runtimes and accelerator toolchains, and engineering reliable real-time computer vision on embedded GPUs.
The confirmed talk is by Antoine Sauvage (EDGEMATRIX), sharing lessons from building and operating Edge AI applications on NVIDIA Jetson platforms. The other two speakers are still to be announced, and the date, schedule and Tokyo venue are provisional. The evening closes with an hour of networking.
About the community
Tokyo AI is a large international AI community based in Tokyo, bringing together engineers, researchers, investors and product people. It runs frequent English-language technical talks, panels and networking evenings connecting Japan's AI builders with the global ecosystem.
#edge-ai#inference#gpu#embedded-systems#computer-vision#ml-systems#gpu-profiling#ai-accelerators#embedded#tech-talks