Cloud AI Acceleration Research Intern (DPU & Infra)

ByteDance; San Jose (CA); Hybrid; USD 60,000 - 80,000; Full time

Job summary

ByteDance is seeking a Cloud Acceleration Research Intern (DPU & AI Infra) for a 2027 start. The internship is based in San Jose within the Technology team.

This PhD-oriented role focuses on research and development of DPU-based cloud acceleration and AI infrastructure, exploring co-design opportunities across hardware and software. You will contribute to optimizing distributed AI training and inference, conduct performance analyses, and build prototypes to evaluate new acceleration techniques.

Responsibilities

• Conduct research and development in DPU-based cloud acceleration and large-scale AI infrastructure.
• Explore software-hardware co-design opportunities for AI/ML infrastructure using DPUs, GPUs, and custom hardware to optimize distributed training and inference.
• Develop new techniques for accelerating distributed AI training and inference, including communication, data movement, resource management, and memory or cache systems.
• Perform end-to-end performance analysis and optimization across hardware, device drivers, OS kernels, communication libraries, and user-space runtimes.
• Build prototypes and evaluate proposed solutions.

Job description

ByteDance is seeking a Cloud Acceleration Research Intern (DPU & AI Infra) for a 2027 start. The internship is based in San Jose within the Technology team.

This PhD-oriented role focuses on research and development of DPU-based cloud acceleration and AI infrastructure, exploring co-design opportunities across hardware and software. You will contribute to optimizing distributed AI training and inference, conduct performance analyses, and build prototypes to evaluate new acceleration techniques.