Senior Inference Engineer, AGI

Highlights

• Own inference for real-time multimodal conversational AI end to end.
• Make frontier-scale speech models run fast enough to feel human in real time.
• Close the loop between live serving and training that improves conversation.

Ideal candidate

Systems-minded engineer who thrives at the science-to-production boundary and obsesses over real-time feel. You enjoy owning ambiguous, high-leverage problems end to end with scientists and hardware partners.

Requirements

• Experience optimizing inference for large-scale multimodal models including attention and KV-cache
• Experience with quantization, speculative decoding, operator fusion and paged KV-cache
• Experience developing high-performance kernels for production serving
• Experience profiling with Nsight Compute/Systems and roofline analysis
• Experience building real-time streaming serving with continuous batching and scheduling
• Experience with multi-GPU inference including tensor parallelism
• Experience building offline rollout and reward-model serving for RL