AI for Atoms: How Periodic Labs is Revolutionizing Materials Engineering with Co-Founder Liam Fedus

So even further back, I was a physics major in undergrad, spent some time doing dark matter research.

Yeah. Why do you think that is? I think it's a great way to think about the world.

It's like very principled, very hard nosed scientists, very careful. And I don't know. I think it's just it's such a incredible field.

You have such high leverage in computer science and AI. Mhmm. And so I think a lot of physicists were seeing that.

Mhmm. Particularly in like high energy physics. After the discovery of the Higgs, I think a lot of high energy physicists were sort of looking for what's next.

Ultimately, it becomes bottlenecked on the new apparatus for pushing the next energy frontier. And I think a lot of physicists were looking at their skillset and looking at the progress elsewhere and saying like, hey, I think I could be a huge contributor elsewhere. Mhmm.

It's just been fascinating to see string theorists and people working on black holes and all sorts of effects moving into AI. Absolutely. Yeah.

then in grad school in physics, was always gravitating towards the machine learning problems. I was looking at particle reconstruction and it's thinking effectively machine learning problems. But it felt if I really wanted to push frontier of machine learning, I should be in computer science.

So ended up at Google Brain, was overlapping with the first year residents there. Absolutely remarkable group of people, remarkable period for Google Brain. I mean, it's the era of when there's the creation of distributed training strategies, mixture of experts, the transformer.

It was a really rich period in that history. And it was a fun Cambrian era where people were really pushing the frontier with just a handful of GPUs, really small collaborations. The field was much much earlier, and I think there was a lot of diversity and entropy in the research, and it was very fun.

2017. Mhmm. So Google brand at that point was still really small, and eventually it was subsumed by DeepMind or combined with DeepMind.

So I was at Google for many years, mostly just doing architecture work. So was really pushing sparsity that allows for more efficient serving of models at scale and just really pushing the scale of what we could do. Towards late twenty twenty two, really became excited about the creation of products.

The technology was getting very compelling, and so I ended up at OpenAI with some other Googlers as well. Mhmm. And what did you work on specifically at OpenAI?

Well, so the goal was we need to come up with some productionization of GPT-four. So OpenAI had GPT four. It was pretrained, and there's some, like, LeRouf post trains on it.

Mhmm. And there's questions about, like, okay, how do we turn this incredibly powerful model into products? And we're all spitballing ideas like writing bot, coding bot, very natural at the time.

Some of our least interesting ideas were a meeting bot. So it would just sit in a Google Meet, take notes, and then send out like to dos after. But John Truman was very opinionated.

He's like, we think we should keep it very general. Let's do a chatbot.

of connecting these systems to the physical world. The opinion that I and others held as periodic was you're not going to see the same kind of acceleration in science and technology unless you start connecting these things to the physical world. Science ultimately isn't sitting in a room thinking really hard.

You have to conduct experiments. You have to learn from them. You have to interface with reality.

And the creation of ChatGPT in late twenty twenty two was a, you know, technology, but it was still far too weak. Like, we couldn't have done periodic on technology of that era. Mhmm.

I think over the next few years past that, we saw ever improving models. We saw reasoning. I think, like, test time inference became really important.

That led to more reliable error correction, more reliable tool use, and we see like the rise of coding agents and other agents. Mhmm. And I think those were foundational technologies necessary to then connect these systems to the physical world.

physics simulations and we have experiment. And, I think exactly as you're pointing out, ML systems are good on the data you've trained them on, on the tasks you've trained them to do. I think sometimes there's this mythology of AGI, ASI, RSI, And I think we see increasingly powerful systems, but they do become limited if they don't have access to the the raw data to actually make informed decisions.

How much how much data do you need? And so I know that, there's some data scale related Yes.

So there is some generalization from the existing models. So we don't need to reproduce a system that can understand and write English or write code. So we're leveraging And are you using open source for that or closed source models or some We use a combination.

Uh-huh. Yeah. So for example, like Periodic spends zero effort on improving coding models.

Mhmm. We're, you know, incredibly impressed by Codex, ClogCode, and so that's been a huge accelerator for the company, but focused our machine learning efforts where the existing frontiers is not sufficiently good for us. I think going back to the data question, we're leveraging, call it order tens of trillions of tokens that went into open source models.

And that's given us very foundational understanding. But once we start moving into specific discovery areas, chemical spaces, we can see a very high level of sample efficiency. So the system isn't starting as like a randomly initialized neural net, it has a strong prior on the world.

So where does that prior come from? What data that informs that? Just general Just like papers, the Internet as you're pointing out.

Yeah. However, that's insufficient. One of the engineers on our team was looking at a reported material property, and it was just sort of extracted values from literature.

It was really interesting to see the reported value spanned many orders of magnitude. And so you train an ML system on that, and it's like, well, the best you can do is model this distribution, but you're no closer to like a ground truth. And that's where experimental data comes in where you now have a grounding in this.

Mhmm. But really important, it's not just like a pool of data. It's this interactive closed loop system that is so powerful.

Once you have the experimental data, you can look through it, you can look for aberrations, you can look for patterns, you can look for consistency with simulation data, with literature, and then that helps drive the next set of experiments.

We have seen internally the greatest advances where we have an abundance of data in some space, and that that has led to the highest rate of acceleration internally. But I think you can think of different levels of generalization. And for systems that are strongly governed by quantum mechanical effects, there is some generalization there.

I see. But if you produce a system that has modeled quantum mechanical objects really accurately, it's not really helping much on fluid dynamics or another level of abstraction.

language models are incredibly powerful. It's a very natural interface, and so we continue to use these. Okay.

But we think about them almost as like an orchestration layer. So that's sort of a a copilot assistant, but also like a system that can direct experiments. And it's orchestrating other specialized models as well.

So we do construct neural nets that are specially designed for atomic systems where there's like some symmetry awareness, and those have much lower latency and they've been like fine tuned for that.

We've treated periodic as our customer zero and seeing how can we transform how this field of science is done. Mhmm. But there's huge opportunities across all of these industries, all these enterprises that are interfacing with the physical world.

People who are bottlenecked by materials engineering, process engineering. And again, those are kind of this like the same natural interfaces where engineers are asking questions about their data, they're trying to find aberrations, they're trying to debug machinery, they're trying to get to a better formulation. It's actually quite universal thing as well.

thinking about us ourselves as an intelligence layer for for these companies. So you can think about system of record, control plane for different experiments and getting to solutions.

Well, I mean, think as you're pointing out, you're going from systems that aren't just writing essays, not just writing software, but to literally generating matter. Mhmm. And I think it has pretty profound implications to semiconductors, aerospace, energy.

And I think it's it's incredibly important for can we increase like the pace of just like the physical development of the world? Mean, we see how quickly the digital realm is changing. Software engineering now looks wildly different than even six months ago.

Mhmm. But I think we see, like, you know, similar opportunities in the physical world. Of course, like, atoms are hard, and so you will have some limits of physics.

But just because atoms are hard doesn't mean there's not an order of magnitude or two to speed up just making sense of huge amounts of data and getting to solutions more quickly. Yeah. So I think what we're trying to do is give humanity this agency for atomic rearrangement synthesis, and we think it's gonna just be a huge accelerator.

The iteration between these groups of people. I mean, this is just irreducibly a multidisciplinary problem. We have physicists and chemists working really closely with some of the top AI researchers in the world, working closely with some of the best engineers in the world.

And this multidisciplinary, like, close collaboration is just absolutely incredible because seeing firsthand how a field can fundamentally change people who have been doing research for, in some cases, decades in a field and now seeing like, oh, under these systems, under intelligent systems, it could look this very different way. And I mean, I use like an analog to machine learning a lot going back to the early Google Brain days where the frontier was pushed forward by a few GPUs and a few people. Now you look at this era where it's really industrialized and there's dozens, hundreds of researchers working together with hundreds of thousands, millions of GPUs dictated and driven by scaling laws.

Everything is about scaling. It's given that predictability. It's allowed us to put huge amounts of capital into this field.

And I think the physical sciences, physical engineering will have a very similar property where we establish these scaling properties and bring that mindset. And so periodic in this field is really thinking about how do we bring much larger scale sets of experiments to bear on this. And intelligence systems have enabled this, automation has enabled this, and you really need both.

An improvement to automation where you can soon become create bottlenecks in intelligence.

are so extraordinarily expensive. And what's interesting is just the compute costs relative to physical infrastructure is actually surprising where, you know, so much money is spent on the compute that the physical infrastructure sometimes is actually lower, but, you know, has very large lead times and there's intrinsic difficulty of having these well calibrated, well functioning physical systems.

team. Are there specific roles that you're actively looking for right now or key things that you really want to hire up? Absolutely.

So on our site, we have decomposed the world into bits and atoms. It's a loose taxonomy, But on bit side, we're really thinking about mid training, pretraining roles from the AI side, always more infrastructure roles. And on atom side, like control engineering, system engineering, but also now thinking too about spanning that with product engineering.

I think one fallacy is thinking about intelligence as a scaler. We've consistently seen these systems have a very odd spikiness, and it's actually possible to architect a system that is world class on some math domain, but then you could do some perturbations to the questions and actually degrade it substantially. So it's like a bad high school student.

And so there's this odd spikiness to these systems. So basically, you can make a system that's like a genius at one thing and not very good at a bunch of other stuff. And I guess the point I was making is those fields can actually be quite adjacent.

So sometimes the generalization can be non intuitive. But one way I think about recursive self improvement is really kind of akin to neural architecture search from, you know, roughly ten years ago. And I think there's a very clear path for software engineering.

So these systems have become so incredibly impressive on this domain as a result of huge amounts of data, really cheap verifiable environments. Like, you know, you can check Unitesco from failing to passing with just a few CPUs. It's basically instantaneous.

There's no domain expertise gap between an AI researcher or software engineer. And obviously, this will become and is becoming a larger contributor to the next generation of the system.

domain caveat to that. Sure. So rolling forward that software engineering self improvement, I think you're going to have a system that can write complete repositories, identify bugs, refactor code, but it doesn't suddenly understand biology.

Sure. Right? It's just like there's a domain gap there in knowledge.

Yeah. But even beyond that, there's sets of strategies done in software engineering that differ from scientific or engineering strategies. So it's You're not operating under It's not like decision making under uncertainty to the same degree.

It's like very verifiable, and that's driven so much of our work. Mhmm. So in that domain, I think it's happening now ish.

Mhmm. And I think we'll see the same thing too for AI research. That's a slower outer loop because now the experiment isn't just checking some unit tests passing, but it's checking what was the scaling property, did this model converge, what's the generalization of the system.

connection of these things to the physical world is going to be so critical because both of those systems are being trained in a closed loop against that domain. So it's a closed loop for doing software engineering, a closed loop for doing AI research, and that's the premise of periodic. Like, we need to have these closed loops of actually doing science, of actually doing engineering.

And these two domains are how I think the rest of the world will go with some delay. And this is again like the foundational technology that we're Super interesting.

goal for Periodic is to generate high quantity, high quality data, diverse data, and automation is assistance to that. So right now, we employ people as well, and we have autonomous parts that are just very reliable. If you had a dextrous humanoid who could wander into an unstructured lab and make sense and follow instructions reliably, that would be a huge accelerator.

Right now, the automation of physical systems is requires a very careful design, and it's slow. But I think with improvements in robotics, it's just going to accelerate this.

It's very simple, very commoditized, not doing a huge amount of innovation on that front.

I mean, of course, robotics. Mhmm. Again, I'm like I'm just so excited about the interface of AI systems with the physical world.

Mhmm. And we're approaching one angle of that, which is science engineering. Mhmm.

And we need that data in order to make those advances. But simply just agency and control of the physical world via robotics is going to be transformative. So I'm very excited about these interface layers.

I think that's gonna be such a massive opportunity. Because, I mean, you know, how many software engineers are there in the world versus people who build, like, the physical world? Mhmm.

And there's just labor shortages everywhere. So, yeah, I think it's gonna be a very interesting decade. Oh, amazing.

Well, thank you so much for joining us today. Yeah. Well, thanks so much.

Was really really good chatting Yeah.