Model Performance Engineer

Job Summary

A company is looking for a Model Performance Engineer to optimize inference performance for AI models on their platform.

Key Responsibilities

• Optimize inference performance by minimizing latency and maximizing throughput
• Experiment continuously to achieve industry-leading performance for various models
• Impact the performance of applications serving millions of users globally

Required Qualifications

• Experience with state-of-the-art inference stacks such as PyTorch, TensorRT, or vLLM
• Open to candidates with any level of experience, including new graduates
• Ability to work in a fast-paced environment and adapt to new challenges
• Willingness to work in-person in New York City or remotely if exceptionally qualified
• Visa sponsorship available for qualified candidates