Senior Staff Inference Performance Engineer (Fully Remote)

D-Matrix; United States; Remote; USD 180,000 - 240,000; Full time

Job summary

d-Matrix is an AI infrastructure company pioneering memory-centric architecture for efficient AI inference. The Inference Cloud team builds deployment and orchestration for d-Matrix's inference services.

You will diagnose performance issues across the serving path, fix or hand off with evidence to owning teams, and build telemetry and playbooks to prevent recurring problems. The role requires deep Linux, NUMA, Kubernetes expertise and Python/C/C++ skills.

Qualifications

• 8+ years performance/systems engineering with profiling expertise.
• Deep Linux, NUMA, Kubernetes, and LLM serving knowledge.
• Proficiency in Python and C/C++.

Responsibilities

• Diagnose performance issues across serving path and isolate root causes.
• Fix directly or hand off with evidence to owning teams.
• Build telemetry and playbooks to prevent recurring problems.

Job description

• The Inference Cloud team builds deployment and orchestration for d-Matrix's inference services.
• You will own end-to-end workload performance, matching bare metal standards.

What You Will Do:

• Diagnose performance issues across serving path and isolate root causes.
• Fix directly or hand off with evidence to owning teams.
• Build telemetry and playbooks to prevent recurring problems.

What You Will Bring:

• 8+ years performance/systems engineering with profiling expertise.
• Deep Linux, NUMA, Kubernetes, and LLM serving knowledge.
• Proficiency in Python and C/C++.

D-Matrix

d-Matrix is an AI infrastructure company that pioneered a memory-centric architecture for efficient AI inference. The culture values respect, collaboration, and direct communication, with humble experts driven by execution.