Senior ML Research Scientist, Perception Models
Twelve LabsVideo Understanding company
Seoul, South KoreaSenior
Data & AI
About the role
TL;DR
Lead research at the intersection of spatiotemporal modeling and multimodal representation learning.
- •As a Senior ML Research Scientist, you will lead research at the intersection of spatiotemporal modeling and multimodal representation learning.
- •Key Responsibilities Research multimodal representations that connect temporal and spatial structure with semantic understanding.
- •Develop models that operate across multiple levels of granularity, from full videos and clips to regions and entities.
- •Design training objectives, datasets, evaluation methods, and large-scale experiments.
- •Use task-specific vision models as supervision or components, and integrate useful signals into general-purpose representations.
- •Partner with research and engineering teams to scale and ship new capabilities.
- •Requirements A strong research track record in computer vision, video understanding, multimodal learning, or a related field.
- •Deep expertise in at least one of the following: video foundation models, self-supervised or contrastive learning, embeddings, and retrieval; or temporal modeling, object-centric learning, detection, segmentation, and tracking.
- •Strong hands-on skills in Python and PyTorch or an equivalent deep learning framework.
- •Proven ability to design and run rigorous experiments at scale.
- •Research impact through publications, production systems, or both.
Required skills
PythonPyTorchComputer VisionNLP
Domain expertise
ai
Benefits & perks
Growth & Tools, Meal & Snack, Wellness & Family
Tech stack
PythonPyTorch