About this role
Sr. Principal Software Engineer at Cerence. Location: Burlington or United States or Europe or Asia or North America. Role: optimizing inference, deploying runtimes, designing quantization Requirements: Proven experience optimizing ML inference in production, deep GPU architecture knowledge, hands-on CUDA kernel development, quantization techniques (INT8/INT4/FP4/FP8/AWQ/GPTQ), and edge/embedded deployment expertise. Category: Software Development Seniority: Senior Level Tools: vLLM, TensorRT‑LLM, llama.cpp, QAIRT, CUDA, AWQ, GPTQ Commitment: Full Time Workplace: Remote Languages: English