below marketmost similar roles pay hereabove market
This role pays more than
92%
of similar roles. Most pay
$174,900–$255,250
— the blue band above.
At the midpoint, this role pays about
$294k
versus about
$215k
for comparable roles.
Based on 240 similar postings.
Employer
About Arm Holdings
Arm Holdings plc is a leading British semiconductor and software design firm, established in 1990 and recognized for developing energy-efficient processor architectures that power nearly all smartphones and a vast range of IoT and computing devices.
Arm Holdings currently has
113 open roles
on FindRole.
Listed pay typically runs
$208,450–$282,050
across 112 roles with salary data.
The Principal System Performance Telemetry Architect joins the Central Technology Group's system architecture team to shape performance telemetry and profiling for AI infrastructure, edge platforms, and physical AI systems. You will architect how performance data is captured, correlated, and exposed across complex SoCs, including CPUs, accelerators, memory, interconnects, and multi-die systems. Daily responsibilities involve defining hardware counters, events, metrics, and traces while driving specifications for workload characterization and bottleneck identification. You will collaborate with SoC architects and software teams to translate debug needs into hardware requirements. Required skills include experience with heterogeneous SoC architectures, performance modeling, and tools like NVIDIA Nsight, Intel VTune, Linux perf, Arm Streamline, and Arm Performance Studio. You will solve technical problems regarding hardware-software observability across firmware, operating systems, and virtualization layers.
Define system-level performance telemetry architecture for hardware counters, events, metrics, and traces across SoCs.
Drive architecture specifications for workload characterization and bottleneck identification across cloud, edge, and AI platforms.
Architect mechanisms to collect and correlate performance telemetry across hardware-software layers.
Translate performance and debug requirements into specific hardware telemetry specifications.
Provide architectural guidance to software and firmware teams for effective hardware telemetry utilization.
Support silicon bring-up, performance validation, and debug activities across multidisciplinary teams.
What we're looking for
Relevant experience in system architecture, performance engineering, performance analysis, profiling, or performance modeling for complex compute platforms.
Strong understanding of heterogeneous SoC architectures, including CPUs, accelerators, memory systems, interconnects, I/O, and multi-die systems.
Experience with hardware performance counters, events, traces, profiling infrastructure, or system-level observability.
Hands-on experience with profiling tools such as NVIDIA Nsight, Intel VTune, Linux perf, Arm Streamline, Arm Performance Studio, or equivalent tools.
Strong understanding of performance analysis methodologies, including workload characterization, bottleneck identification, latency and bandwidth analysis, and power-performance optimization.
Understanding of how hardware telemetry is exposed and consumed across firmware, operating systems, virtualization layers, runtimes, and developer tools.
Experience collaborating with cross-functional teams with strong communication and technical leadership skills.
Experience with silicon bring-up, performance validation, debug, or hardware-software performance correlation (preferred).
Linux
Cloud-Native
Containers
Orchestration
AI Systems
Benchmarking
Profiling
NUMA
Memory Bandwidth
Arm Neoverse
Cross-ISA
Hardware Performance Counters
System Telemetry
MLCommons