"ultimately sciences driven against experiment in the real world. And so that's what we're doing with periodic labs. We're taking these precursor technologies and we're saying okay if you care about ad..."

"so Liam was working on LLMs and they were going really well. Um, and I was not using LLMs, but I was noticing that LM are becoming more and more impactful in my work. So one way it was becoming impact..."

"found periodic. >> Well, let's take a second to talk about what periodic is and what does it do. So periodic labs is a frontier AI research lab that's trying to use LLMs to advance physics and chemist..."

"for folks who might not be familiar to explain what you guys mean by a lab that will verify RL in the real world. Can you talk a little bit about how experiments work? How does how do how are AI model..."

"This lays a foundation, but ultimately science is driven against experiment in the real world. And so that's what we're doing with periodic labs. We're taking these precursor technologies and we're sa..."

"been so focused on basically putting together like this N of one team like these group of physicists, chemists, uh um simulation experts and some of the best machine learning researchers in the world ..."

"know naively the current pre-training and post-training uh sort of pipelines at most of the frontier labs won't just eventually crack physics as well. Why why is is this idea of physical verification ..."

"case of periodic I think you mentioned the first bel lines you guys are making are towards superc conductivity and magnetism right what is it about those domains that make them good candidates for the..."

"important for us to do the loop, but then create this repeatable process like how do you go from subdomain to subdomain? And there's really interesting questions about how well do the ML systems gener..."

">> So I I can see the argument for why um cracking room temperature superconductivity from an experimental basis is is extraordinary valuable for humanity. But you guys are building a startup. And um ..."

"maybe perhaps just being in Silicon Valley we you know we really think about like computer oriented work everything is digital everything is bits but there's so many industries Like we were kind of ta..."

"deeper on the that in practice sort of a day in the life of a periodic uh team member where let's say half the team is this roughly right about half the team are ML scientists with machine learning ba..."

"closely with each other. Um, so one of the things we're seeing is the physics and the chemists need to figure out how to teach the LLM how to reason about these things because I think the frontier AI ..."

"we're making towards our internal superconductivity or physics goals actually is making our LLMs much better at serving our customers who are doing very similar workflows. Yeah, I think just culture, ..."

"taking a lot of the research that's going on at the company and deploying that out in the real world, the kinds of customers that we've talked about, space, defense, advanced manufacturing, these are ..."

"with pre-training, they're familiar with post- training, but in the periodic context, what does mid training mean? >> Yeah, sorry for the lingo. So, I think this this term came up uh years ago where i..."

"that a lot of the fundamental research in the physical sciences that uh we're talking about here physics, chemistry, biology has historically been done at university labs. Um is there a role at all th..."

"interested in joining periodic what are you guys looking for? First off, someone deeply curious. Uh someone who really wants to understand the machine learning, the science at a deeper level, who want..."