Why it matters
LLM routing is moving into the platform layer — gateways that meter token usage and pick the best-fit model per prompt; for platform and infra engineers, this is what governing AI traffic looks like in practice on Kubernetes.
The tokenmaxxing angle
Pure tokenmaxxing territory: Solo.io demos agentgateway monitoring token usage and preventing runaway AI spend, and DigitalOcean's Inference Router auto-routes prompts to the best-fit LLM — token metering and model routing as live infrastructure.
From the organizers
Talks from Shane O'Donnell (VP Engineering, Solo.io) on Gateway API plus agentgateway for token-usage governance, and Prasad Prahbu (DigitalOcean) on its Inference Router; in-person only at 90 Broadway, Cambridge, with Solo.io sponsoring food.