Architecture Energy Modeling Engineering Intern - 2027

Responsibilities

• Work with architects and performance architects to develop an energy-efficient GPU.
• Develop methodologies and workflows to select and run a wide variety of workloads to train models using ML and/or statistical techniques.
• Develop methodologies to improve the accuracy of energy models under various constraints, such as, process, timing, floorplan and layout.
• Correlate the predicted energy from models created at different stages of the design cycle, with the goal of bridging early estimates to silicon.
• Develop tools to debug energy inefficiencies observed in various workloads run on silicon, RTL and architectural simulators. Work with architects to fix the identified energy inefficiencies.
• Work with performance, verification and emulation methodology and infrastructure development teams to integrate energy models into their platforms.
• Prototype new architectural features, create an energy model, and analyze the system impact.
• What we need to see
• Pursuing MS degree
• Strong coding skills, preferably in Python, C++.

Languages