Deep Learning Performance Intern - 2027
At a glance
Junior AI Research role at NVIDIA. China, Shanghai · full-time.
Pay not stated AI Research salary
aimlroles summary, based on the employer's posting.
What you'll do
- Benchmark machine-learning workloads across GPU and NPU designs, then analyze how their performance compares
- Build models to forecast how deep-learning workloads will perform on different architectures
- Trace slowdowns to hardware, software or system bottlenecks and recommend ways to improve performance
- Test hardware features in deep-learning applications to understand their capabilities
What you bring
- A bachelor’s, master’s or doctoral degree in a relevant area such as computer science, electrical engineering or mathematics
- Familiarity with GPU or accelerator-based deep-learning platforms and their software stacks
- A solid grounding in computer architecture
- Experience with system architecture design and performance optimization
Who this fits
This internship suits someone interested in measuring and improving deep-learning performance across hardware architectures. Familiarity with LLM or generative AI algorithms, kernel optimization and machine-learning frameworks is also relevant.
From the employer
NVIDIA is developing processor and system architectures that accelerate deep learning on edge devices, workstations, and data center GPUs for a variety of applications including automotive, robotics, large language models and AI generative models. We are looking for an expert deep learning system performance architect to join our deep learning modelling, performance optimization, projections, and analysis effort. In this position, you will have the chance to optimize deep learning hardware and software architecture and make the significant impact in a dynamic technology focused company.
What you’ll be doing:
• Benchmark and analyze performance of various machine learning/deep learning algorithms on different GPU/NPU architectures
• Develop performance models and provide performance projections for deep learning workloads on different architectures
• Identify architecture, software and system performance bottlenecks and propose optimizations
• Explore new features and hardware capabilities on deep learning applications
What we need to see:
• BSc. MS or PhD in relevant discipline (CS, EE, Math, etc.,)
• Be familiar with GPU or Accelerator-based deep learning platform and software stack
• A strong background in computer architecture
• Be familiar with LLM or generative AI deep learning algorithms and kernel optimizations
• Experience in system architecture design and performance optimization
• Familiar with machine learning and deep learning frameworks