We are seeking a highly skilled Power and Performance Developer and Analyst with a strong focus on tool development, coding, and scripting as well as data analysis to join our machine learning data center team. In this role, you will be responsible for driving the automation of power and performance analyses, including developing the necessary infrastructure and frameworks for efficient data collection and analysis. You will collaborate with architecture, hardware, firmware, and software teams to define key metrics and ensure that data-driven insights lead to significant optimizations in data center performance and power efficiency. This role also involves developing software to interface with firmware and hardware components for real-time power and performance data collection.

Key job responsibilities
* Develop and automate tools and frameworks for running training and inference workloads, as well as collecting power and performance metrics.
* Use programming and scripting languages (C/C++, Python, Bash, etc.) to create efficient workflows for automating the execution of complex workloads and data collection processes.
* Perform in-depth data analysis of collected power and performance metrics, identifying key trends, bottlenecks, and opportunities for optimization.
* Work closely with cross-functional teams (architecture, hardware, firmware, software) to define data collection strategies, select appropriate metrics, and develop automated analysis tools.
* Build and maintain interactive dashboards and data visualization tools to communicate insights from performance and power data clearly and effectively to technical and non-technical stakeholders.
* Contribute to the creation of end-to-end automated pipelines for continuous power and performance testing, monitoring, and reporting.
* Design and implement data validation processes to ensure the accuracy and integrity of collected power and performance metrics across various workloads.
* Perform root cause analysis of performance and power inefficiencies, using custom scripts and tools to debug and optimize system-level performance.
* Optimize data collection and analysis processes to improve efficiency, scalability, and accuracy in high-performance environments.
* Collaborate with the team to build a comprehensive test suite for validating power and performance benchmarks across hardware and software configurations.
* Stay current on emerging tools, technologies, and best practices in power and performance analysis, contributing to the evolution of our tool set and frameworks.