Principal Power & Performance Validation Engineer

About the role

The role centers on defining and executing the validation strategy for Graphcore's AI silicon and systems to ensure strict adherence to power, thermal, and performance targets. You will own the technical direction for power and performance validation activities across the product lifecycle, from bring-up through to end-of-life scenarios. This position requires deep expertise in characterizing complex AI compute hardware under demanding workloads and environmental conditions. You will act as the primary authority on power and performance behavior, translating architectural specifications into concrete validation plans and measurable results. The position is embedded within the Architecture and Validation team, where your analysis directly influences product readiness and system reliability. You will be responsible for mentoring engineers while driving the technical rigor required for high-integrity silicon validation. Success in this role depends on your ability to manage intricate hardware and software interactions to deliver predictable and optimized platform behavior. You will partner closely with architecture, firmware, software, and system teams to resolve critical issues that span multiple domains.

Key facts

What you'll do

• Direct the validation strategy for thermal, power, and performance characterization of AI compute platforms across multiple generations.
• Manage, mentor, and develop a team of validation engineers while coordinating project priorities and resolving resource constraints.
• Create comprehensive validation plans that address stress testing, workload scenarios, corner cases, and long-duration system reliability.
• Oversee post-silicon bring-up and characterization for both silicon and complete systems to ensure compliance with specifications.
• Validate critical subsystems including memory controllers, CPU clusters, interconnect fabrics, and high-speed I/O interfaces.
• Build and maintain automation frameworks and regression infrastructure using Python to scale validation coverage efficiently.
• Design benchmarks and parameter sweeps that expose system optimization opportunities and potential performance regressions.
• Partner with firmware, RTL, software, and architecture teams to debug complex technical issues that span multiple layers of the stack.
• Establish clear reporting methodologies, metrics, and pass/fail criteria for validation data and performance results.
• Communicate project status, technical findings, and risks to engineering leadership to support informed decision-making.
• Define test strategies and structured validation methodologies that ensure thorough coverage of power, performance, and thermal scenarios.
• Collaborate with systems and application teams to validate real-world workloads and ensure that silicon meets deployment requirements.
• Drive the development of characterization datasets that inform power and thermal models used throughout the product lifecycle.
• Implement traceable validation processes that support audits, compliance, and continuous improvement initiatives.
• Lead investigations into performance anomalies, power spikes, and thermal events to identify root causes and corrective actions.

Requirements

• Demonstrate proven experience in silicon characterization, system performance engineering, or power/performance validation within a hardware or semiconductor environment.
• Show previous experience in a technical leadership or management role where you guided engineers through complex validation challenges.
• Possess comprehensive knowledge of SoC architecture, including high-speed I/O, interconnects, memory subsystems, and CPU clusters.
• Show proficiency in Linux systems and low-level performance analysis using standard tools and interfaces.
• Demonstrate advanced Python skills for data analysis, orchestration, and automation of validation workflows.
• Show familiarity with profiling and benchmarking tools such as iperf, perf, fio, and stress-ng in real-world validation scenarios.
• Exhibit ability to debug complex interactions between hardware and software across multiple layers of the stack.
• Demonstrate experience defining test strategies and structured validation methodologies that scale across large teams and projects.
• Show ability to manage cross-functional initiatives across global teams, aligning objectives and resolving dependencies.
• Demonstrate strong written and verbal communication skills for documenting validation results and presenting findings to technical leadership.
• Show commitment to maintaining high standards of accuracy, repeatability, and transparency in all validation activities.
• Demonstrate experience working within regulated or quality-conscious environments where documentation and traceability are critical.
• Show willingness to collaborate closely with academic and research teams where applicable to leverage cutting-edge methodologies.
• Demonstrate ownership of process improvements that enhance validation efficiency, coverage, and reliability over time.

Nice to have

• Background in high-performance compute, GPU, or AI accelerator architectures and their associated validation challenges.
• Experience with simulation or pre-silicon modeling environments that complement post-silicon validation activities.
• Understanding of silicon characterization techniques and advanced power management technologies such as dynamic voltage and frequency scaling.
• C/C++ programming skills for developing benchmarks, low-level system utilities, or performance analysis tools.
• Familiarity with telemetry systems, hardware instrumentation, and real-time data collection frameworks.
• Experience working with large-scale or high core-count compute systems, including multi-socket and distributed environments.
• Experience scaling engineering teams and establishing validation best practices across multiple projects.

Practical notes

• This is a full-time position based in Bengaluru, India.
• Travel may be required to Graphcore offices or customer sites as needed.
• Visa sponsorship may be considered for eligible candidates.
• Applicants must meet the requirements outlined in the job description.
• Only candidates who closely match the outlined profile will be contacted for further discussion.