Principal Applied Scientist, ML Codesign - Amazon.com Services LLC

Key Responsibilities

• Define joint optimization strategies for model compression and silicon architecture to meet performance, power, and area targets.
• Set technical specifications that propagate through the model, compiler, runtime, and silicon layers.
• Collaborate with cross‑functional teams to validate compression algorithms on real hardware and refine design trade‑offs.
• Drive innovation in quantization, large‑model distillation, and other compression techniques for production workloads.
• Mentor and lead a multidisciplinary team of researchers and engineers.

Requirements

• Ph.D. or equivalent experience in Machine Learning, Computer Architecture, or related field.
• Deep expertise in model compression, quantization, and distillation techniques.
• Strong understanding of silicon design, compiler construction, and edge/cloud inference pipelines.
• Proven track record of leading technical initiatives and delivering production‑ready solutions.
• Excellent communication skills and ability to influence across multiple engineering domains.