V
Overview
Veeda AI builds high-fidelity, scalable simulated environments powered by world models to support Physical AI and robotics. Its core product creates an internal multimodal simulation that embodied agents, like robots, can learn from through interaction in a safe virtual space. The approach uses multimodal foundation world models that integrate vision, depth, physics, and other data to simulate how the world behaves, enabling faster and safer learning at scale. The goal is to provide a robust simulation-based infrastructure for the AI, robotics, and embodied intelligence ecosystem, reducing bottlenecks in learning and deployment.
Work here?Claim your company
Simplify's Take
What believers are saying
- Veeda AI raised over $90 million in August 2026, a landmark Canadian seed.
- It posted eight open roles by September 14, 2026, signaling aggressive hiring.
- Khosla Ventures and Radical Ventures provide elite distribution for follow-on financing.
What critics are saying
- Veeda AI still ships no product, leaving valuation tied to research execution.
- Its $90 million seed buys compute, but world-model training burns capital rapidly.
- Nvidia, Google DeepMind, and Toyota-backed labs crowd the same robotics simulation race.
What makes Veeda AI unique
- Sanja Fidler led Nvidia's Toronto Spatial Intelligence Lab for eight years.
- Veeda AI targets multimodal world models for Physical AI, not generic foundation models.
- The company spans Toronto, Zurich, and California, recruiting simulation and robotics specialists.
Help us improve and share your feedback! Did you find this helpful?