Catalyst Lab
Catalyst Lab works end-to-end on high-value, high-uncertainty problems from foundational theory to production code
Catalyst Lab advances foundational ideas and builds them into end-to-end systems.
We work on ambitious technical challenges that benefit from tight iteration between research and execution. That means developing new ideas, building working systems, and learning from real-world constraints along the way.
Who we are
Catalyst Lab brings together researchers, engineers, technical program managers, partners, and AI-enabled tools to move from early invention to durable capability.
Our work often spans the full arc of innovation. In some efforts, we pursue new foundational advances. In others, we focus on building, integrating, and refining systems. In many cases, progress comes from doing both together.
How we work
Our teams are small, highly technical, and cross-disciplinary. We work quickly, learn through building, and stay grounded in real user needs and operational reality.
Catalyst Lab is also a place to explore new ways of working in the age of AI. As teams increasingly build with AI and agents, we experiment with how collaboration, development, and coordination can evolve.
Areas of focus
AI memory and data systems
How AI systems store, retrieve, update, and use knowledge reliably over time
Security tools and sensors
How we detect risk, surface weak signals, and build systems that improve security posture and response
Dense-signal systems
How we turn complex signals such as audio, acoustics, power, energy, or brain data into usable models and real-world tools
Sociotechnical AI
How AI systems interact with people, institutions, and communities in ways that are effective, trustworthy, and workable in practice
Cryptography
How we build and apply advanced cryptographic methods to protect systems, data, and communications in changing threat environments