Advance multimodal foundation world models at Veeda AI through generative modeling, PyTorch, and distributed training. Work on Physical AI research.

As a Member of Technical Staff, you will design and scale predictive models across visual and spatial modalities. You will develop multimodal learning methods, improve sequence modeling, and explore post-training techniques. Your experiments and evaluations will guide modeling and data decisions.

Key Responsibilities:

- Design and train generative and predictive models
- Develop multimodal learning and conditioning methods
- Improve consistency, robustness, and generalization
- Explore post-training and model optimization techniques
- Conduct evaluations and controlled experiments

Requirements:

- PhD in Computer Science, Engineering, or related field
- Experience training machine learning foundation models end to end
- Solid PyTorch and distributed training experience
- Ability to design and analyze machine learning experiments
- Experience developing evaluations revealing model limitations

Apply research expertise in PyTorch, sequence modeling, and evaluation to Veeda AI's multimodal Physical AI world models.

📌 World Model Research Scientist (Toronto)
🏢 Veeda
📍 Toronto