16 days ago
San Jose, CA, USASenior
Base Salary
$200k - $400k/yr
Responsibilities
- Design and train large-scale video foundation models using internet-scale and robot-collected datasets.
- Develop pretraining methods that capture temporal dynamics, motion, and object interaction.
- Build transferable representations for perception, tracking, prediction, and control.
- Explore transformer-based and diffusion-based architectures for video understanding and generation.
- Implement efficient video data pipelines and distributed training strategies for high-throughput ingestion and large-scale model training.
- Optimize model performance across compute, memory, and training-efficiency constraints.
- Collaborate with generative modeling, agent, and robot learning teams to integrate pretrained models into the autonomy stack.
- Design evaluation frameworks and benchmarks for temporal understanding, prediction quality, and generalization.
Requirements
- Experience training large-scale models on video data or other high-dimensional sequential modalities.
- Strong understanding of deep learning architectures for video, vision, or multimodal systems.
- Experience with large-scale pretraining, dataset curation, training dynamics, and scaling laws.
- Proficiency in Python and deep learning frameworks such as PyTorch.
- Experience with distributed training systems and large GPU clusters.
- Strong experimental rigor and ability to iterate quickly on model design and training strategies.
- Solid software engineering skills and ability to build scalable, reliable systems.
- Ability to work independently and drive ambiguous, high-impact research directions.
- Preferred qualifications include experience with frontier video or multimodal foundation models, video diffusion, autoregressive video modeling, world models, large-scale video dataset construction and filtering, robotics or embodied AI, egocentric video, and publications in machine learning, computer vision, or multimodal AI.
Benefits
- Full-time role requiring five days per week of in-office collaboration in San Jose, California.
- The total compensation package may include additional components and benefits depending on the specific role.
Categories
AI ResearchML Engineering
About Figure AI
Figure AI builds general‑purpose autonomous humanoid robots and the AI software that powers them, targeting tasks in industrial, commercial, and eventually residential settings. The privately held company monetizes through robotics hardware and related software/services aimed at enterprises seeking automation. Founded in 2022 and headquartered in San Jose, California, it has announced a partnership with BMW Manufacturing to pilot humanoid robots in automotive production.