Hi, I'm Stephanie Crater.

I'm a PhD candidate in bioengineering at University of California, Berkeley and University of California, San Francisco studying how the brain allocates limited computation during decision-making.

How do intelligent systems decide when to think, what to think about, and how long to think? I study how the brain solves this problem by recording hippocampal theta sequences and dopamine signals (biological implementations of forward search and TD learning) in behaving animals to understand how the brain simulates and evaluates potential futures under resource constraints. I'm broadly interested in which computations are worth doing at all — which trajectories to simulate, which outcomes to learn from, and which memories to store or replay — and what this tells us about intelligence in both biological and artificial systems.

Before studying memory in the brain, I worked on memory management in the JVM. See below for some artifacts from my past life as a software engineer at Microsoft, building and optimizing the Microsoft Build of OpenJDK.

I graduated from Duke University in 2021 with degrees in biomedical engineering, electrical engineering, and German. My undergraduate research focused on improving the accuracy of connectomes generated via diffusion MRI, resulting in a first-author cover article in NeuroImage.