​Why Attend?

​Join the world’s leading AI engineers, researchers, and enthusiasts in the heart of Paris for a 2-day intensive hackathon and workshop, focused on optimizing LLM inference performance using AMD’s powerful MI300X GPUs. Whether you're a systems engineer, an ML researcher, an AI enthusiast or a student this is your chance to:

• ​Get hands-on with vLLM, SGLang & kernel optimization

• ​Showcase your skills in LLM pipeline tuning

• ​Compete for performance-based prizes

• ​Learn from AMD, Hugging Face, Meta & more! 

• ​Get a chance to win a €599 Developer Pass to the RAISE Summit, happening from July 8–9! (Conditions apply – only 50 passes available)

​Seats are limited. We encourage you to register in advance to secure your spot.

​Agenda:

​Day 1:

• ​9:30 AM – Doors Open

• ​10:00 AM – Opening Talk by Ramine Roane on "What’s New with AMD’s Developer Initiatives?"

• ​10:15 AM – Hackathon Introduction

• ​10:30 AM – Workshop 1: Optimizing Model Serving with vLLM – SEUNG ROK JUNG & Felix Marty- See Details Below

• ​11:15 AM – Workshop 2: Developing Optimized Kernels – Matej Sirovatka See Details Below

• ​12:00 PM – Lunch + Networking + Hackathon Team Formation

• ​1:00 PM – Hackathon Kickoff+ Overview on Rules, Resources & Support

• ​5:45 PM – Partner Talk @ Station F

• ​6:00 PM – Dinner & Networking

​Day 2

• ​8:00 AM – Doors Open + Breakfast Networking

• ​8:45 AM – Partner Talk: Hugging Face- Remi Ouzan

• ​9:00 AM – 2:00 PM – Hackathon Continues

• ​12:00 PM – 1:00 PM – Lunch

• ​2:00 PM – 3:00 PM – Final Benchmark Runs

  • ​Standardized script

  • ​Teams submit performance logs

• ​3:45 PM – Partner Talk: AI Infra: What’s Powering Llama and What’s Next? by Davide Italiano, Meta

• ​4:00 PM – 5:00 PM – Top 10 Team Presentations (5 min each)

• ​5:00 PM – 6:00 PM – Judging & Deliberation

• ​5:30 PM – 8:00 PM – Awards + Happy Hour

​Workshop Details:

​1. Optimizing Model Serving with vLLM 

​An advanced user, instructor-led, hands-on GPU workshop equips you with tips and techniques to maximize the performance of at-scale vLLM deployment on AMD GPUs. Join this exciting course, led by AMD architects. Learn about the challenges in production-scale model serving, optimize performance with deployable kernels, and apply them to an LLM to boost efficiency, gaining confidence in enhanced performance and TCO benefits

​2. How to Develop Optimized Kernels 

​An advanced user, instructor-led, hands-on GPU workshop with an introduction to AMD GPU architecture and developing optimized kernels. This course is led by GPU Mode team. Explore GPU architecture, learn kernel development basics start with a simple kernel example, and join a GPU Mode kernel project to build practical skills.

​Hackathon Details

​Theme: End to end inference pipeline performance optimization of an opensource LLM on AMD MI300X GPUs

​Prize Money

• ​1st Prize- €2,000 Per Team

• ​2nd Prize- €1,500 Per Team

• ​3rd Prize- €1,000 Per Team

• ​A GPU Credit of 100 hours Per Team

​A team can consist of either a single individual or up to a maximum of three people.

​To qualify, should deliver at least 5% higher performance compared to base reference.

​Lookout for more details soon

​