Monte Carlo tree search

So what is different about this than just a generally a, a general agentic system that can sort of optimize notebooks?

Uh, right now it's a, uh, it's essentially its own in modern twenty twenty-six parlance, we actually worked on this in twenty-four and twenty-five, but in modern, modern parlance, it's kind of a specialized harness that, uh, runs an algorithm, uh, which in, in for ERA was Monte Carlo tree search.

So essentially it's keeping, um, hundreds or thousands of possible instances of notebooks, and then it selects one, and I can explain how it selects one, and it decides, "Well, how can I, what, what can I do?" Gemini asks itself, "What can I do to make that, uh, notebook be better?" And then it will make a new one and test it and then put it back into the candidate pool.

So you can imagine the candidate pool is actually tree structured because it's, um, uh, you know, every candidate possibly has some children.