Head of AI Accelerator Performance & Telemetry

The Consensus; San Jose (CA); On-site; USD 180,000 - 260,000; Full time

Benefits offered by this job

Medical/dental/vision

Housing subsidy

Relocation assistance

Wellness programs

Daily meals

Compute budget

Job summary

Etched is hiring a Head of Performance Visibility to define performance metrics and abstractions for next-generation AI accelerator systems. This role spans hardware and software, designing cross-layer telemetry, counters, and distributed reasoning across nodes, racks, and clusters.

You will shape scalable tooling and insights for engineers debugging large-scale AI systems. We operate in a fully in-person team in San Jose (Santana Row) and seek someone who can drive architecture across hardware

Qualifications

• Experience building profiling or observability systems.
• Ability to translate raw hardware signals into scalable telemetry and analysis infrastructure.
• Experience correlating time-series events across distributed systems.
• Deep systems programming experience (C++ or Rust).
• Experience designing distributed tracing or performance modeling frameworks.

Responsibilities

• Define the architectural approach for collecting and structuring telemetry across CPUs, drivers, interconnects, and multiple accelerators
• Design scalable models for correlating performance events across device and host boundaries
• Develop mechanisms to align hardware counters, runtime activity, and workload semantics across model-layer execution into coherent insight
• Implement time synchronization and trace-alignment strategies across multi-device systems
• Build tools that identify bottlenecks among multi-accelerator workloads across chips within hosts
• Contribute to analysis engines and developer-facing tooling that transform raw telemetry into intuitive insight

Skills

Systems programming

Distributed systems

C++/Rust

Telemetry/observability

Time-series analysis

Job description

Etched is hiring a Head of Performance Visibility to define performance metrics and abstractions for next-generation AI accelerator systems. This role spans hardware and software, designing cross-layer telemetry, counters, and distributed reasoning across nodes, racks, and clusters.

You will shape scalable tooling and insights for engineers debugging large-scale AI systems. We operate in a fully in-person team in San Jose (Santana Row) and seek someone who can drive architecture across hardware