← All open roles
ML Engineer, 4D Action Segmentation
Build the parsing pipeline that turns continuous 4D hand-mocap streams into clean, verified, training-ready tasks. No annotation team.
What you'll do
- Develop and improve action-segmentation models that parse 4D point-cloud streams from the glove.
- Iterate on architectures (transformer-based, point-cloud, hybrid) and characterize where each fails.
- Build labeling and verification tools that close the loop between collectors and the dataset.
- Define evaluation suites that catch regressions before they hit production.
- Partner with simulation, robotics, and product teams to make the parsed output land where it matters.
What we're looking for
- MS or PhD in CS, robotics, or related field - or equivalent demonstrated impact.
- 3+ years working with sequence models, point clouds, or video understanding.
- Strong Python; PyTorch fluency.
- Track record of moving models from research code to something a teammate can rely on.
- Comfortable working at the boundary between learned models and structured pipelines.
Nice to have
- Published work in action segmentation, 4D understanding, or temporal models.
- Experience with HOI4D, EgoExo4D, or comparable datasets.
- Familiarity with point-cloud architectures (PointNet++, P4Transformer, PPTr).