Applied AI Research Engineer

Job Summary

Owning inference research in a full-time capacity, the Applied AI Research Engineer will focus on low-level inference optimization and customer workload enhancements while working remotely or from New York or San Francisco.

Key responsibilities

• Conduct low-level inference optimizations, including speculative decoding, quantization, and memory management
• Collaborate directly with customers to optimize production workloads and implement improvements to the platform
• Exercise a high level of autonomy to identify and guide the development of high-impact features for the inference platform

Required qualifications

• 3+ years of experience in systems or research related to LLM inference
• Deep understanding of LLM serving, including kernel and scheduler functionalities
• Proven history of shipping products or research used in production environments
• Strong enthusiasm for collaborating with customers and engaging with developer tools and cloud-native technologies
• Experience with GPU programming and frameworks like Torch/PyTorch