Data Center Digital Twin CFD Development Intern
Spring 2024
Confirmed live in the last 24 hours
Designer & manufacturer of computer chips & graphics processors
Company Overview
NVIDIA is on a mission to solve the world's most stimulating technology problems – in industries ranging from gaming to scientific exploration.
AI & Machine Learning
Company Stage
Grant
Total Funding
$4.2B
Founded
1993
Headquarters
Santa Clara, California
Growth & Insights
Headcount
6 month growth
↑ 13%1 year growth
↑ 5%2 year growth
↑ 21%Locations
Santa Clara, CA, USA
Experience Level
Intern
Requirements
- Currently pursing an MS or PhD in Mechanical Engineering, Aerospace Engineering, Computer Sciences, or a closely related field
- Shown experience in Omniverse
- Strong experience with Python programming language
- Familiarity with complex techniques for solving Navier Stokes equations, turbulence modeling and associated BCs such as using FFD, RANS, LES… modeling techniques
- Shown excellence in thermal-hydraulic and fluid dynamics modeling using powerful CFD tools. Moreover, basic understanding of the thermodynamics of data centers and CFD
- Excellent problem-solving skills and attention to detail
Responsibilities
- Build high-fidelity CFD AI surrogate models by applying artificial intelligence and machine learning in order to predict near-real-time data center CFD simulations; we work with designated data center equipment vendors to create an AI/ML assisted CFD simulations algorithms for real-time control
- Craft the opportunity to implement the most complex pixel-computing algorithms in constrained environments, such as embedded systems based on Nvidia's Tegra SOC and DT applications built in Nvidia's Omniverse platform
- Analyze and integrate data from a variety of sources, such as IoT sensors, monitoring systems, and network devices, to calibrate simulation results for AI/ML accelerated CFD simulations plugged into Omniverse DTs in data centers
- Model data center facilities using computational fluid dynamics and flow networks, from chip to chiller, for creating digital twins. Model Nvidia's high heat density servers, racks, and PODs for air, single phase, two phase, and immersion cooling; analyze the impact of various scenarios on data center operations with the predictive CFD tool embedded in a digital twin, including capacity planning, energy efficiency optimization, and fault prediction
- Engage in multi-functional collaboration with data center operations, IT, engineering, Omniverse digital twin modeling team, AI/ML experts, and others to gather requirements, tools, and techniques