Senior Staff Inference Performance Engineer (AI)

Location: Santa Clara, California or remote within the United States; hybrid

Salary: $175,000–$265,000 per year, plus equity and bonus opportunities.

Company

hirify.global develops software and hardware for generative AI inference, including Corsair accelerators and cloud-based inference services.

What you will do

• Own end-to-end performance across orchestration, scheduling, networking, autoscaling, inference serving, runtimes, interconnects, operating systems, NUMA, and accelerators.
• Profile workloads, isolate bottlenecks to specific layers and owners, implement fixes or workarounds, and provide actionable cross-team handoffs.
• Protect throughput, latency, and stability targets by closing performance gaps between orchestrated workloads and bare-metal deployments.
• Assess fleet-scale routing, placement, and shared-resource contention across heterogeneous inference architectures.
• Capture CPU, memory, NUMA, and network requirements in orchestration configuration so performant deployment is the default.
• Build telemetry, benchmarks, and investigation playbooks while mentoring engineers in a profiling-first methodology.

Requirements

• 8+ years of hands-on performance or systems engineering experience, including cross-layer regression analysis.
• Deep knowledge of Linux internals, NUMA, CPU pinning and cpusets, PCIe, memory and cache behavior, and network paths.
• Experience running performance-sensitive workloads on Kubernetes and identifying the cost of scheduler defaults, network overlays, and orchestration abstractions.
• Working knowledge of LLM inference, including batching, KV-cache, prefill/decode, quantization, and speculative decoding.
• Proficiency in Python and C/C++ for instrumentation, debugging, and cross-stack code changes.
• Clear technical writing, sound ownership judgment, and disciplined use of AI and agentic tools validated against real data.

Nice to have

• Experience with vLLM, SGLang, TensorRT-LLM, speculative decoding, disaggregated serving, or heterogeneous serving architectures.
• Experience tuning KV-cache management, request and batch scheduling, or weight prefetch and residency.
• GPU programming and profiling experience with CUDA, Nsight, or NCCL.
• Experience with pre-production hardware, Rust, or startup environments.

Culture & Benefits

• Collaborative and inclusive environment emphasizing respect, humility, direct communication, and execution.
• Medical, dental, and vision coverage.
• 401(k) and an employee rewards package focused on employee and family wellbeing.
• Equity and performance-based bonus opportunities.