MTS Inference Performance Visibility
EtchedVisit Etched website
Etched is an AI-hardware company building rack-scale frontier inference clusters.
San Jose, United States
Funding history
About Etched
Etched co-designs chips, racks, software, and manufacturing systems for efficient inference of frontier AI models, targeting throughput, latency, cost, and power efficiency across prefill and decode workloads.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will design and build a performance-analysis suite for ML accelerators. You will collect and correlate hardware, driver, host, PCIe, and system performance data; create analysis modules and visualizations; identify bottlenecks; and guide engineers in using the tooling for optimization.
Requirements
- Proficiency in C++ or Rust
- Deep understanding of computer architecture, memory hierarchies, and PCIe
- Experience with low-level performance analysis, profiling, and bottleneck identification
- Experience with performance analysis tools
- Experience working close to hardware, performance counters, or device drivers
Responsibilities
- Lead the design and architecture of a performance-analysis suite
- Capture performance data from custom ML accelerator hardware
- Implement tracing for host-side API calls and system-level events
- Correlate performance events across CPUs, drivers, PCIe, accelerators, and hosts
- Build analysis modules that identify performance limiters
- Develop visualizations of performance characteristics and bottlenecks
- Define tool requirements and provide performance-analysis guidance
Benefits
- Medical, dental, and vision coverage
- $500 per month credit for waiving medical benefits
- $2.5k per month housing subsidy for employees living within walking distance of the office
- Relocation support to San Jose
- Wellness benefits covering fitness and mental health
- Daily lunch and dinner
- Unlimited compute budget subject to ROI justification