Rethinking AI and Computing

Artificial Intelligence is not only transforming how we live and revolutionizing all sectors, from healthcare to finance, but it’s even reshaping computing itself. Models are growing, workflows are becoming more complex, and hardware comes in many forms. But while AI is advancing rapidly, today’s computing systems remain rigid and centralized, unable to keep pace with this fast-changing landscape.

At UT Austin’s InfraAI Center, we see this challenge as an opportunity. We are building AI infrastructure and systems that adapt, learn, and scale — from cutting-edge accelerators to edge devices and massive datacenters. Our mission is to tackle today’s pressing needs while reimagining how future systems are designed, managed, and optimized.

Our vision is built on two complementary pillars:

1. Nimble AI Stacks: We are creating flexible, high-performing AI software stacks that evolve with models and hardware. By embedding intelligent learning into the stack itself, we make AI infrastructure more efficient, adaptable, and accessible to everyone—democratizing innovation beyond well-resourced organizations.

2. Intelligent Infrastructures: We are designing AI-driven systems that manage themselves. By developing new building blocks and abstractions, these systems adapt in real time, optimize performance, and maintain safety across cloud, edge, and consumer devices—reducing the need for constant human intervention.

Together, these pillars define a future where AI and computing co-evolve, delivering intelligent, resilient, and widely accessible systems ready for the next generation of innovation.