Get Out of the Model's Way — Kevin Hou, Google DeepMind
AI Engineer · · 19:01
Kevin Hou discusses designing AI products that improve as models improve, separating agent orchestration from the IDE, and the scale of resources used in an OS-kernel demonstration.
the one thing that has stood above all other lessons that I've talked, talked about is the idea of scaling with intelligence. This means that as the model gets better, so should your product. And the frontier edge of whatever model you are serving should be apparent inside of your user's product experience.
We believe that the IDE is to the agent manager what the debugger was to the IDE. You don't always need a debugger, but it definitely is helpful to have it if you need to go a layer beneath and go one step deeper into that abstraction stack.
An OS-kernel demonstration used extensive agent resources
Original excerpt
it took ninety-three sub-agents over the course of twelve hours, made fifteen thousand requests, two billion tokens, and it was under a thousand dollars, which was one of the really cool aspects of this project.
nafyi.