Research Scientist, Multi-Modal Human Understanding

What you'll do

• Design multimodal architectures and develop Vision-Language Models for human-centric understanding.
• Build video foundation models for temporal reasoning, action synthesis, and video generation.
• Run experiments, evaluate benchmarks, analyze failures, and contribute across the research lifecycle.

What they ask for

• Bachelor's degree in Computer Science, Computer Engineering, or a relevant technical field, or equivalent practical experience; degree must be completed before joining.
• 2+ years of multimodal AI research experience, including hands-on work with Vision-Language Models, video understanding, or human-centric AI systems.
• 2+ years implementing and training large-scale neural networks, including transformer-based architectures.
• Experience with Python, multimodal model evaluation, and developing or fine-tuning Vision-Language Models.

Nice to have

• Experience with video foundation models, temporal transformers, or large-scale video pretraining.
• Published multimodal AI research at venues such as CVPR, ICCV, or NeurIPS; experience with generative models for human synthesis.

Pay

$122,000/year to $181,000/year base salary

Tools

Vision-Language Models (VLMs), PyTorch, Python, transformer-based architectures, diffusion models, GANs, autoregressive models

AI focus

Research and develop multimodal AI for human understanding and synthesis, including video and motion generation.