GECI Int.

Diagnose and triage hardware and network incidents on GPU compute clusters. Collect hardware logs and identify failed components for resolution or RMA.

The main mission is to massively accelerate a sparse calculation engine by porting critical algorithms from Rust/C++ to CUDA/GPU, aiming for performance gains of 100x to 1000x. Responsibilities include developing performant Rust/CUDA wrappers, optimizing sparse matrix operations for GPU, designing GPU memory management strategies, and collaborating with the engineering team on integration.