BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//events//Events//EN
CALSCALE:GREGORIAN
X-WR-CALNAME:Make Your GPU Earn Its Keep: A Kernel Engineering Deep Dive
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-PUBLISHED-TTL:PT1H
BEGIN:VEVENT
UID:event-ryji8tgh@ontown.app
DTSTAMP:20261110T093354Z
DTSTART:20261130T170000Z
SUMMARY:Make Your GPU Earn Its Keep: A Kernel Engineering Deep Dive
LOCATION:San Francisco\, United States
DESCRIPTION:AI Performance Engineering Meetup (San Francisco\, Global)\nMon
 day\, November 30\, 2026 · 9:00–10:00 AM PST · Online\nZoom link: http
 s://us02web.zoom.us/j/82308186562\nA powerful GPU can spend plenty of time
  waiting. What is it waiting for—and how do you find out?\nTalk #0: Welc
 ome and Meetup Updates\nby Chris Fregly and Antje Barth\nTalk #1: From “
 Why Is This Slow?” to a Faster Kernel\nby Dr. Rajat “Raj” Dandekar\,
  Co-founder of Vizuara\nJoin Raj for a focused tour of the engineering beh
 ind fast AI kernels\, drawing on Vizuara’s Kernel Engineering course on 
 Maven. We’ll connect the roofline model to CUDA execution and the GPU me
 mory hierarchy\, then look at how profiling guides optimizations such as c
 oalesced memory access\, tiling\, and tensor-core use. Matrix multiplicati
 on and attention provide context for seeing how those ideas fit together i
 n modern AI workloads\, including FlashAttention.\nThe goal is a practical
  way to decide what to measure\, which bottleneck matters\, and whether an
  optimization helped. The session is dedicated to Raj’s talk\, with time
  for audience questions.\nWho it’s for: Engineers comfortable with Pytho
 n or basic PyTorch. Prior CUDA or C++ experience is not required.\nBio: Dr
 . Rajat Dandekar is a co-founder of Vizuara and a Kernel Engineering cours
 e instructor. He holds a PhD in Mechanical Engineering from Purdue Univers
 ity and BTech and MTech degrees from IIT Madras.\nBased on: Kernel Enginee
 ring: From CUDA Foundations to Frontier AI Kernels (https://maven.com/raja
 t-dandekar/kernel-engineering)\nScheduling note: November 30 is the propos
 ed new date\; speaker confirmation is pending.\nZoom link: https://us02web
 .zoom.us/j/82308186562\nRelated Links\nGithub Repo: http://github.com/cfre
 gly/ai-performance-engineering/\nO’Reilly Book: https://www.amazon.com/S
 ystems-Performance-Engineering-Optimizing-Algorithms/dp/B0F47689K8/\nYouTu
 be: https://www.youtube.com/@AIPerformanceEngineering\nGenerative AI Free 
 Course on DeepLearning.ai: https://bit.ly/gllm
URL:https://ontown.app/e/ryji8tgh-make-your-gpu-earn-its-keep-a-kernel-engi
 neering-deep-dive/
END:VEVENT
END:VCALENDAR