Senior Researcher, Efficient AI

What you'll do

• Develop and evaluate new algorithmic approaches for AI serving to optimize latency and throughput.
• Design and experimentally validate endpoint configuration policies, including batching and routing strategies.
• Conduct hardware-aware optimization by aligning serving algorithms with accelerator capabilities and attention innovations.
• Build experimental prototypes and large-scale measurements to drive research ideas toward production readiness.
• Publish research findings and contribute to open-source systems for AI inference serving.

What we're looking for

• Doctorate or equivalent experience in a relevant field.
• 3+ years of experience designing and optimizing efficient inference systems.
• Expertise in machine learning frameworks like PyTorch, TensorFlow, and serving frameworks such as vLLM, Triton Inference Server.
• Proficiency in GPU programming with CUDA, ROCm, and other GPU optimization tools.
• Strong knowledge of algorithmic optimization, parallel computing, and request orchestration under strict SLO constraints.
• Research impact through publications and patents, along with hands-on experience delivering research ideas to production.