ML Engineer, Model Compression
Core
Optimizing trained AI models for speed, memory efficiency, and reliability on specific hardware (GPUs, NPUs, TPUs) for deployment in client environments and QuTwo OS.
Role type
Senior IC ML Engineer (Model Compression & Optimization)
Builds
Efficient, production-ready inference systems for industrial AI applications
Domain
Quantum computing systems, Industrial AI, Edge/On-device ML
Deliverable
production ML models
Required skills
Model compression (quantization, pruning, knowledge distillation), Profiling (memory, bandwidth, compute), Hardware-aware optimization, Open-weight model families, Domain adaptation harnesses, Python, C/C++
Preferred skills
Custom kernel development (CUDA, Triton, HIP/ROCm, Metal), On-device/edge ML, Concurrency/parallel execution optimization, Model serving at scale, Synthetic data generation
Technologies
CUDA, Triton, HIP/ROCm, Metal, NVIDIA Jetson, QuTwo OS
Responsibilities
Take trained models from research and clients and optimize them for real hardware limits; Profile models to identify bottlenecks in memory, bandwidth, and compute; Optimize models for specific accelerators; Build scaffolding for domain adaptation; Deploy optimized models to client environments and QuTwo OS.
Seniority
Senior, hands-on IC
