Application Software Engineer, Inference

📋 Description

• Develop highly reliable, high-throughput inference systems across SpaceX
• Architect scalable distributed infra for model serving: load balancing, auto-scaling
• Optimize latency under production workloads with GPU kernels, quantization
• Build high-concurrency serving with uptime, low tail latency, observability
• Own end-to-end components such as routing, SDKs, rate limiting, scaling
• Benchmark and accelerate inference engines (SGLang, vLLM, TensorRT-LLM)

🎯 Requirements

• Bachelor's degree in CS/engineering/math or 2+ years software experience
• Experience designing and maintaining scalable distributed systems
• 1+ year full-stack or backend development with production systems
• 1+ year Rust or C++
• Experience with LLM inference engines and serving frameworks (SGLang, vLLM, Triton, TensorRT-LLM)
• Deep low-level systems programming: GPU kernels, batching, caching, quantization

🎁 Benefits

• Medical, vision, and dental coverage; 401(k); disability and life insurance
• Paid time off: 3 weeks vacation and 10+ holidays; paid sick leave
• Employee stock purchase plan and long-term incentives
• Parental leave and other discounts and perks