Research at the intersection of machine learning, data systems and agentic reasoning.
What We Do
Data Systems
Learned components for document databases: query optimization, cardinality estimation, index selection, and schema design for systems where data doesn't fit neatly into rows and columns.
Generative Models
Neural architectures that work natively with hierarchical, semi-structured data — learning to predict, generate, and embed JSON documents without flattening them into tables first.
AI-Assisted Research
Building tools and methods for AI-augmented scientific research: experiment tracking designed for AI agents, reproducible workflows, and infrastructure that lets researchers and AI collaborate effectively.
How We Work
We practice what we research. Our lab runs on automated pipelines that triage issues, route tasks to AI agents, and track experiments with minimal human intervention. The same infrastructure we build as research tools powers the lab itself.
Every process that can be delegated to an agent is. Humans drive prioritization, design decisions, and interpretation — the mechanical work of running experiments, reviewing code, and managing workflows is handled by the systems we build. This isn't a future aspiration; it's how we operate today.