Software Architecture Engineer Intern

VisLab (an Ambarella Inc. company); Kentucky; On-site; USD 28,000 - 41,000; Full time

Job summary

Ambarella invites self-motivated Master’s students to join our AI Vision Processors internship near completion of their degree. You will work on software models for hardware co-processors, exploring memory bandwidth and pipeline latency, and create visualization tools to depict architecture behavior.

We value experience in C++, Python, CUDA, and familiarity with PyTorch, with opportunities to present findings at the program’s end.

Qualifications

• Pursuing a master’s degree with strong programming foundations.
• Interest in computer architecture, VLSI design, image processing, and video compression.
• Familiarity with modern programming languages and ability to deliver on time.

Responsibilities

• Develop software models of hardware co-processors and memory bandwidth.
• Explore architecture trade-offs between software and hardware.
• Prepare final presentation summarizing findings for ongoing projects.

Skills

C++

Python

Computer architecture

Image processing

CUDA

PyTorch

Education

Master's degree in progress

Tools

PyTorch

CUDA

Job description

AI Vision Processors For Edge Applications

Our solutions make cameras smarter by extracting valuable data from high-resolution video streams.

Job Description

Ambarella has some exciting internship opportunities for self-motivated and creative students near completing of their master's degree. We are looking for talented software programmers who possess a strong interest in computer architecture, VLSI design, image processing, video compression, and computer vision.

Architecture interns study hardware and software trade-offs present in the latest version of our chips and explore uses of both to solve real-world problems. Interns may develop software models of important behaviors in hardware co-processors such as memory bandwidth and pipeline latency or develop visualization tools to illustrate architecture behavior. Interns produce support tools to enable software to make efficient use of hardware co-processors. A final presentation at the end of the internships describes their findings and allows other team members to use their work for ongoing projects.

The internship program requires experience in modern programming languages and computer architecture. C++, C, Python, and CUDA are our popular tools. Exposure to common neural networks frameworks such as PyTorch helps. We expect that interns will be able to pick up missing skills quickly, deliver their assignments on time, and present their work at the end of the program.