All jobs
All jobs
The role studies how model architecture, memory movement, and data-center constraints affect throughput, latency, and total cost of AI inference, using hardware/software co-design. Interns research accelerators, CPUs, memory, and interconnects, then build models or benchmarks and recommend what to build or optimize. Candidates should be pursuing a PhD with a systems, architecture, or ML background, a GPA of 3.0 or higher, and C/C++ programming skills.
HQ Belgium·Hybrid