Senior Power Analysis and Optimization Engineer AI LLM Systems

Senior Power Analysis and Optimization Engineer, AI-LLM Systems

NVIDIA is seeking a Senior Power Architecture & Optimization Engineer to advance energy efficiency in GPUs and SoCs using advanced analytics and AI, including specialized LLMs for power analysis. Join us at the forefront of accelerated computing, where AI and LLMs are shaping the future of power analysis and optimization, providing a key differentiator for our next-generation products. As a crucial member of this effort, you will blend power architecture expertise with machine learning, reinforcement learning, data analytics, and large language models (LLMs). Your work will pioneer new methods for modeling, analyzing, and optimizing energy consumption at both full-chip and unit levels. What You’ll Be Doing

• Analyze full-chip and unit-level power using internal and industry-standard RTL and gate-level power tools, translating data into concrete design and architectural improvements.
• Develop and productionize power-aware models and flows, incorporating ML/RL techniques for anomaly detection, dynamic power management, and design-space exploration.
• Design and train new LLMs to interpret complex power data, propose root causes and fixes, and recommend architectural optimizations for power.
• Perform comparative power analysis across workloads, products, and design options to identify trends and anomalies not apparent from first principles.
• Collaborate closely with Architects, Performance, Software, ASIC Design, and Physical Design teams to interpret power data, root-cause power bugs, and drive solutions.
• Prototype and evaluate new architectural features in Verilog, with a keen focus on power and energy implications.
• Automate and scale flows (Python/Perl/C++), defining new pipelines to accelerate power anomaly detection and integrate power data, AI models, and design decisions.
• Apply AI to power optimization by building and deploying data-driven models (ML, RL, data analytics, custom LLMs) to tune power-efficient configurations and policies.

What We Need To See

• MS (or equivalent experience) and 5 years of experience OR PhD + 3 years of experience in EE/CE/CS or related fields.
• Strong understanding of energy consumption, power estimation, data movement, and low-power design principles.
• Familiarity with Verilog and ASIC design principles, with hands-on experience in tools like PowerArtist, PrimePower/PrimePower RTL, RTL Architect, or similar.
• Solid coding and automation skills, preferably in Python, Perl, and C++.
• Experience or strong interest in machine learning, reinforcement learning, and data analytics, ideally applied to EDA, architecture, or system-level optimization.
• Interest or experience in building and using LLMs or other foundation models as engineering copilots, particularly for EDA/power/architecture workflows.
• Excellent communication and collaboration skills for effective cross-functional teamwork.
• A genuine desire to integrate data-driven, AI-assisted decision-making into power architecture and influence the energy profile of NVIDIA’s future products.

Key skills/competency

• Power Analysis
• Optimization
• AI/ML
• LLMs
• Verilog
• Python
• ASIC Design
• Data Analytics
• RTL Design
• Energy Efficiency