Jobs

Modal is seeking a Member of Technical Staff for Research, Inference, to join their team building the new infrastructure layer for AI. This hands-on role involves leading end-to-end inference research, focusing on optimizing cost per token and tail latency for customer workloads. You will work on cutting-edge techniques like speculative decoding, quantization, and KV-cache management, collaborating with customers and external research labs. The ideal candidate has a strong background in LLM inference, fluency in the LLM serving stack, and a proven track record of shipping impactful research or systems.