LLM Inference Runtime Architect for AI Accelerator

United States Digital Space LLC is hiring a production-grade LLM inference runtime engineer to join the Hardware AI team. You will design and optimize the end-to-end runtime that maps frontier models onto our AI accelerator, balancing latency, throughput, and hardware utilization.

Work spans model architecture, systems software, and silicon interfaces, with emphasis on reliability, observability, and scalable performance in production environments.