Research Scientist - Intern (Video and Edge, Multimedia Lab) - 2027 Start (PhD)

What you'll do

• Develop video understanding and quality assessment methods, including temporal reasoning, event localization, and analysis of visual degradations.
• Build datasets and benchmarks; evaluate model accuracy, robustness, and generalization.
• Explore efficient adaptation and inference of multimodal models and assess prototypes through experiments and, where applicable, production validation or A/B testing.

What they ask for

• Currently pursuing a PhD in Computer Science, Computer Engineering, Electrical Engineering, or a related technical discipline.
• Strong foundations in machine learning and computer vision, with research or project experience in at least one listed area.
• Proficiency in Python and hands-on experience with PyTorch or TensorFlow.
• Familiarity with Transformer architectures, vision-language models, or MLLMs, and hands-on experience with model training, fine-tuning, inference, or evaluation.

Nice to have

• Experience in long-video understanding, temporal grounding, quality assessment, or perceptual modeling.
• Familiarity with video enhancement and compression, Linux, OpenCV, FFmpeg, distributed data processing, GPU-based training, or C/C++.

Tools

Python, PyTorch, TensorFlow, Transformer, MLLMs, LoRA, Linux, OpenCV

AI focus

Research multimodal models for video understanding, temporal reasoning, perceptual quality assessment, and efficient model adaptation and inference; connect semantic and user-preference insights to video enhancement and compression.