Power Architecture Engineer

Responsibilities

• Join the NVIDIA Power Architecture Group to architect, develop, and verify end-to-end power aspects for NVIDIA's family of products.
• Contribute to power analysis by architecting, developing, and verifying power estimation models and tools.
• Architect and design system-level power features to optimize dynamic and leakage power dissipation across various use cases.
• Perform power verification, including structural, functional, and power-aware verification, by developing test plans, testcases, and test bench components.
• Validate the effectiveness of power features on silicon and conduct studies to drive Performance/Watt improvements.
• Drive innovation in energy efficiency and help build advanced power management solutions.

Requirements

• B.Tech./M.Tech degree with 1+ years of experience in power-related domains (Power analysis, Power Design, Power Aware Verification, UPF methodologies, or Power Correlation).
• Strong fundamentals in power, including transistor-level leakage and dynamic characteristics of VLSI circuits.
• Strong fundamentals in digital design and Verilog.
• Familiarity with low power design techniques such as multi-VT, Clock gating, Power gating, Voltage Islands, and DVFS.
• Knowledge of power intent formats such as UPF or CPF.
• Experience with static power check tools (VCLP/CLP) and dynamic power verification tools (VCS-NLP or equivalent).

Preferred Qualifications

• Strong debugging and problem-solving skills.
• Proficiency in Python with experience in object-oriented programming and design.
• Exposure to power analysis tools such as PTPX or EPS.
• Experience with lab setups for power measurements (e.g., oscilloscope/DAQ) and the ability to analyze board-level power issues.

About the Company

NVIDIA is a global leader in accelerated computing. Our invention of the GPU sparked the growth of the PC gaming market, redefined modern computer graphics, and revolutionized parallel computing. Today, we are driving the future of artificial intelligence through highly scalable and massively parallel computation horsepower.