Principal Telemetry Architect - System Performance for AI

Arm; Raleigh (NC); Hybrid; USD 250,000 - 338,000; Full time

Job summary

Arm is seeking a Principal System Performance Telemetry Architect to shape the performance telemetry architecture for AI infrastructure, edge platforms, and physical AI systems. You will define hardware counters, events, and traces, and guide software and tooling teams for effective profiling across SoCs.

This role sits in Arm's Central Technology Group, collaborating with engineers across disciplines. You will drive workload characterization, bottleneck analysis, and power-performance

Qualifications

• Experience in system architecture, performance engineering, profiling, or performance modeling for complex compute platforms.
• Strong understanding of heterogeneous SoC architectures including CPUs, accelerators, memory, interconnects, I/O, and multi-die systems.
• Experience with hardware performance counters, events, traces, profiling infrastructure, or system-level observability.
• Hands-on profiling with Nsight, VTune, Linux perf, Arm Streamline, or Arm Performance Studio.
• Understanding of how telemetry is exposed across firmware, OS, virtualization, runtimes, and developer tools.
• Experience collaborating with cross-functional teams and providing technical leadership.

Responsibilities

• Define system-level performance telemetry architecture, including hardware counters, events, metrics, traces, and data collection across CPUs, accelerators, memory, interconnects, I/O, and multi-die SoCs.
• Drive architecture specifications and performance analysis methodologies for workload characterization, bottleneck identification, and system optimization across Arm-based cloud, edge, and physical AI platforms.
• Architect mechanisms to collect, correlate, and expose performance telemetry across SoC components and hardware-software layers for performance analysis and debugging.
• Collaborate with SoC architects, performance modeling and analysis engineers, and profiling teams to translate performance and debug needs into hardware telemetry requirements.
• Provide architectural guidance to software, firmware, and tooling teams to enable effective use of hardware telemetry for profiling, software optimization, and platform deployment.
• Support silicon bring-up, performance validation, and debug while providing technical leadership across multidisciplinary architecture, hardware, and software teams.

Skills

System architecture

Performance engineering

Telemetry requirements

Profiling tools

Cross-functional leadership

Communication

Tools

NVIDIA Nsight

Intel VTune

Linux perf

Arm Streamline

Arm Performance Studio

Job description

Arm is seeking a Principal System Performance Telemetry Architect to shape the performance telemetry architecture for AI infrastructure, edge platforms, and physical AI systems. You will define hardware counters, events, and traces, and guide software and tooling teams for effective profiling across SoCs.

This role sits in Arm's Central Technology Group, collaborating with engineers across disciplines. You will drive workload characterization, bottleneck analysis, and power-performance