What are some of the learnings from that world that were transferable to the robotics work that you're doing right now?

I mean, like you said, some of it is the approach with which you see a problem and how you go about solving it.

Certainly one of the things that I've learned with each one of them, and if you notice in each one of them, I was there trying to change the way the current community was solving the problem.

So even when I started working on speech recognition, one of the first things that I introduced was how to use deep learning for keyword spotting, which at the time was done in a totally different way.

So I think in most of these cases, I try to go at it from How can I learn this from data in a way that feels robust and resilient to noise? And also, how do I try to solve this without necessarily evaluating against the current techniques, but not just trying to do an incremental improvement on them? So it's perfectly OK to sort of start from scratch and remove the existing approach before anything.

before evaluating against it.

So in speech recognition, by the time that I was in that team, we were rewriting everything.

And I assume that that requires like... maybe like a significantly smaller model or a significantly smaller data set at least?

Maybe we can also talk about the arc to sort of get here and sort of obviously this is the second Gemini robotics model, but is there like other things that are like worth, like obviously Google's been doing a lot of robotic stuff, DeepMind maybe, I actually don't know, maybe also add a bunch of robotic stuff, but can talk about any of sort of the research and the arc to get sort of this launch moment.

Yeah, I mean, it's been a long path.

I think many of us have been here also for a while and have seen all these steps through.

But yeah, I mean, we've always, from the beginning, like I said, we're really thinking about general purpose robotics.

Before, I think the field really realized that that was feasible.

And so we've done a few iterations where we've brought robots sort of techniques that now are table stakes for the community.

So for example, we first introduced like reinforcement learning to learning simulation, how to control whole robots, right? So you see a lot of robots today like dancing and doing acrobatics.

Any advice, and maybe I'm misremembering this, but I feel like we do have a bunch of customers who use the ER models who actually like... aren't robotics companies and they just happen to be in one of these like domains for like video audio spatial understanding or something like that i don't know if that's like a suggested path for for folks but i assume it's like those domains of like video spatial understanding where like the er model is like better on a bunch of these core benchmarks