Staff Power and Performance Validation Engineer
Graphcore is a SoftBank-owned AI-compute company developing AI processors, systems, and software for machine-learning workloads.
Maintainer signals as of 9/23/2026
Funding history
About Graphcore
Graphcore develops Intelligence Processing Units (IPUs) and the Poplar SDK for building and running machine-learning applications. It continues operating under the Graphcore name as a wholly owned SoftBank Group subsidiary.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will define and lead validation strategies for power, thermal, and performance characterization. You will conduct post-silicon bring-up, develop Python automation and regression infrastructure, execute benchmark experiments, analyze performance data, debug hardware and software interactions, and communicate technical findings and recommendations.
Requirements
- Power and performance validation
- Silicon characterization
- System performance engineering
- SoC architecture
- CPU
- Memory
- Interconnect
- High-speed I/O
- Linux
- Low-level performance analysis
- Python
- Benchmarking
- Profiling
- stress-ng
- fio
- perf
- iperf
- Hardware and software debugging
- Validation methodology
- Workload modeling
- Data analysis
Responsibilities
- Define validation strategies for power, thermal, and performance characterization
- Plan and execute validation across projects and milestones
- Develop validation plans for functional, stress, workload, and corner-case scenarios
- Lead post-silicon bring-up and characterization
- Validate CPU, memory, interconnect, and high-speed I/O subsystems
- Develop Python automation, regression infrastructure, and reporting tools
- Design and execute benchmark workloads and performance experiments
- Collaborate to debug and resolve hardware and software issues
- Define validation metrics and pass/fail criteria
- Develop workload generators and micro-benchmarks
- Analyze power, thermal, and performance data
- Communicate findings, risks, and recommendations