London PyTorch #24: Infrastructure for AI

Join us on February 25th for talks from engineers at Nebius, Fireworks AI, Together AI and Cerebras - four companies working on distinct approaches to AI infrastructure, from cloud-native training platforms to custom silicon.

Topics:

• Nebius - Tips and tricks for Training High-Performance Speculators by Filipp Fisin
• Fireworks AI - LLM evaluations and closing the benchmark-to-production gap by Shaunak Godbole
• Cerebras - Technology underpinning the fastest inference by Dmitry Gaynullin
• Together AI - Road to 5M Sequence Length: Breaking Memory Barriers in Context Parallelism by Max Ryabinin

If you’d like to give a talk or become a partner for one of the upcoming meetups, message https://www.linkedin.com/in/fedorshabashev/ or https://www.linkedin.com/in/kirzharov/

In the meantime, you can watch the previously recorded talks:
https://youtube.com/@londonpytorchmeetup

Please subscribe to our channel! ❤️