Multimodal AI Research Assistant: Video Understanding

SUTD (Singapore University of Technology & Design); Singapore; On-site; SGD 42,000 - 54,000; Full time

Job summary

The MEVER lab at the Singapore University of Technology and Design (SUTD) invites a motivated Research Assistant to contribute to multimodal AI research, including understanding and reasoning across modalities and developing models for social good.

The successful candidate will work on online understanding of long-context videos, streaming processing, and maintaining temporal context, with a strong emphasis on PyTorch, experimentation, and rigorous evaluation.

Qualifications

• Strong background in machine learning and deep neural networks.
• Experience with PyTorch and strong programming skills.
• Strong interest in multimodal AI, video understanding, and multimodal foundation models.
• Experience in computer vision, natural language processing, or related areas is desirable.
• Ability to independently implement, experiment with, and evaluate machine learning methods.

Responsibilities

• Develop methods for efficient and effective online understanding of long-context videos.
• Incrementally processing streaming video and maintaining relevant temporal context over extended durations.
• Collaborate with researchers to evaluate and compare models and methods.

Skills

Machine learning

Deep neural networks

Programming

Multimodal AI

Video understanding

Analytical skills

Research skills

Education

Bachelor's degree in Computer Science, Artificial Intelligence, or a closely related field

Tools

PyTorch

Job description

The MEVER lab at the Singapore University of Technology and Design (SUTD) invites a motivated Research Assistant to contribute to multimodal AI research, including understanding and reasoning across modalities and developing models for social good.

The successful candidate will work on online understanding of long-context videos, streaming processing, and maintaining temporal context, with a strong emphasis on PyTorch, experimentation, and rigorous evaluation.