Research Scientist, Video Understanding & World Models

A Research Scientist, Video Understanding at Mecka AI leads the end-to-end video understanding agenda by training large-scale video representation and video-language models on the company's robotics dataset. The role involves designing model architectures, running self-supervised pretraining, and transforming research checkpoints into production signals for robotic systems. Candidates will work closely with engineering teams to ensure their models integrate seamlessly into real-world execution pipelines.