We pursue two connected lines of research to build state-of-the-art,
commercially deployable systems.
Our hypothesis
A research program connecting learning, episodic memory, and explicit search to build more powerful adaptive intelligence.
Read Intelligence Is SearchWe pursue it on two tracks
Track 1
Translate neuroscience & AI research into computational principles for cognition, reasoning, and intelligence.
Track 2
Engineer these principles into state-of-the-art LLM systems that maximize intelligence per FLOP.
Articles
We identify a key principle for identifying where search and planning should occur in robotics, instantiate it with per-skill policies and world models composed at test time, and show novel compositions, inferred subgoals, and replanning without composite demonstrations.
Perturbing a small fraction of routing decisions in frozen MoE reasoning models—a little noise helped, a lot hurt, and the tuning curve suggests a new search surface inside the model.
We hold model weights and prompts fixed, and explore everything else. Our research discovers levers within LLMs that improve reasoning at test time, guided by computational principles drawn from neuroscience.
Dynamic Compute Allocation improves agent performance by deciding, at each turn, where additional computation is worth spending.
AI today encodes intelligence into weights and retrieves it at inference. Hard problems demand more: active reasoning that balances accuracy with efficiency, and knows when to think harder.