Dr. Qichao Hu

Yeah, so that's where we still have human scientists coming in to train this.

So think of the database as just like a dictionary, right?

You still need a person to know, okay, what letter, what word do I look up?

So...

That's where the intelligence comes in.

For each domain, we have about 50 human scientists to teach the model.

For example, we still use the frontier models like the GPT-5 and the Gemini.

And then those are not specifically trained in these domains.

They're very general.

So we would have about a team of 50 scientists.

domain scientists and then teach the frontier models, for example, about batteries, about pesticides, about cosmetics.

In each domain, here are the things that you should look for.

For example, in the battery case, to have a high temperature, stable cycle life, you need the molecule to have this particular structure.

So when you go through the entire database of molecules, look for these structures and then look for melting point, boiling points within a certain range.

Look for energies within a certain range.

So the human scientists would actually teach the frontier model these domain specific knowledge.

And then this intelligence would go look for the corresponding molecules in that database.

So for now, we are using an agentic LEM, and then it's a combination of GPT-5 and then Gemini.

And so all the domain-specific constraints and then the information we would teach this agent, and then the agent will look for it in the database.

So cool.