Modal

The era of actually open AI is here. We’ve spent the past year helping leading organizations deploy open models and inference engines in production at scale.

Hosted by [Charles Frye](https://x.com/charles_irl), this live session will walk you through:

1.  The three types of LLM workloads: offline, online and semi-online.
2.  The challenges engineers face and our recommended solutions to control cost, latency, and throughput
3.  How you can implement those solutions on [our cloud platform](https://modal.com/)


High Performance LLM Inference in Production

Coding agents actually work. At Ramp, over half of merged PRs were written by their background AI agent.

Hosted by [Rebecka Storm](https://www.linkedin.com/in/rebeckastorm/) and [Paul Butler](https://www.linkedin.com/in/paulgb/), this live session will dive into how AI agents code at scale. We’ll walk through:

1.  How customers like Ramp and Lovable use Modal sandboxes today
2.  Security, networking and scaffolding considerations for your coding agent
3.  Deploying Modal sandboxes at scale


Inside Modal Sandboxes: How agents code at scale

Speakers

Paul Butler

Member of Technical Staff

Rebecka Storm

Product Lead

Replays

High Performance LLM Inference in Production