AI Meetup San Francisco - 12/10/2024

Tech Talk: Finetune LLMs with Metaflow cost-efficiently, from 1b to 405b
Speaker: Eddie Mattia (Outerbounds)
Abstract: Imagine a world where fine-tuning is no hassle - just another seamless step in your workflow, (almost) no matter the model size. Accessing GPUs for fine-tuning shouldn't be a hassle either. And you should be able to do all of this securely and privately in your cloud account. Over the past 18 months, open-source Metaflow, originally started and still actively developed at Netflix, has added a host of new features. It now helps you fine-tune LLMs in your own environment, using popular tools and tapping into compute resources across clouds. In this talk, we will walk you through a practical example of how you can do this at home today, highlighting the latest and greatest in Metaflow.

Tech Talk: Crossing the AI Prototype Chasm
Speaker: Matt Wyman (Co-Founder and CEO at Okareo)
Abstract: In this session, we will dive deep into the intricate factors that influence agent behavior and provide actionable insights and practical techniques to fine-tune it for optimal performance.