Research engineering

GPU systems through tensor contractions

Building GPU systems through tensor contractions, with a focus on hardware understanding, memory movement, and algebraic structure.

About

I build around tensor contractions as a way to understand how computation maps onto hardware.

tensor-energy-workspace is the foundation for exploring tensor algebra, contraction order, data layout, memory traffic, and GPU execution.

The goal is to connect algebraic structure and hardware intuition, not just raw throughput.

We study GPU computation from a structural perspective. Tensor contractions are used as a lens for understanding how algebraic structure maps onto GPU hardware, data layout, and memory movement.

Tensor algebra & contraction order
Data layout & memory traffic
Energy measurements
GPU execution & occupancy

Project

Open-source foundation for understanding GPU structure through tensor algebra.

tensor-energy-workspace

Foundation for exploring tensor algebra, contraction order, data layout, memory traffic, energy measurements, and GPU execution.

PythonCUDATensor networksEnergy

Contact

Open to collaborations on GPU systems, tensor-network methods, and high-performance computing.