Can we ask about
your evals?

We’re Rory and Malthe. We just moved to SF for Entrepreneurs First.

We have demo day in December and are figuring out what to build in the ai eval space.

So we’re talking to teams about evals: how you test your agents, what’s working, what’s still a pain, and which tools do you currently use.

We are curios what you think about agent evals. Don't feel like you need to prep anything.

Rather text?

iMessage / SMS

A little about us

Previously product at Zopa and Aspora, where I led the stablecoin rail and AI payments product.

Previously founding engineer at Paper Instruments, working on agent evals, post-training and RL infrastructure.

Coauthor of DiligenceBench, an equity-research agent evaluation.