Hack RadarMaking LLMs faster: An intro to ML compilers
- beginner-friendly
- in-person
- free
waterloo's getting a talk on why llms are slow and how compilers fix it, speaker's vijay goyal from openai. free entry, beginner-friendly, and if you rsvp on luma and actually show up to check in/out you get a free chatgpt sub. stay after 4:30 for studio hour if you want 1-on-1 time with the team.
the full listing
How do we make LLMs run faster? 👀 The Waterloo OpenAI Campus Leads are teaming up with WAT.ai to bring you an introduction to ML compilers with Vijay Goyal! We’ll explore why compilers exist and how they help models run faster. This specialized talk is for both technical and non-technical students. We’ll get everyone up to speed on the basics before diving into deeper technical content and real examples. No prior compiler knowledge is needed, and we’ll share ways to keep exploring with OpenAI tools afterward! Before you arrive: Please have your ticket QR code ready. You’ll need it to enter STC 0060! Free ChatGPT subscriptions: To be eligible, please RSVP on Luma and complete both workshop check-in and check-out in person. We’ll share instructions during the workshop. Book Giveaway: Some attendees will receive a free copy of 100 Chats for College Students, featuring ideas from students for using ChatGPT in studying, careers, and everyday life! Want to put what you’ve learned into practice? Stay for Studio Hour from 4:30-5:30 PM to ask follow-up questions and get 1-on-1 support using OpenAI tools for your own projects and ideas.
source: Luma