Banseog note · 2026-10-26
A world model lab building the 'virtual reality' robots learn in
Current stage · from public events · 2026-11-14
- R&D · prototype
- Pilot · first production
- Scaling production
- Operations · field rollout
- Going global · localizing
- M&A · integration
- Restructuring
R&D · prototype Secured early funding to develop a multimodal world model
- 2026-08-19 A transaction adviser confirmed Veeda's funding and its intended use for developing a physical AI world model. Source
Roles often seen at this stage Robot software, AI & vision · Controls & embedded
Companies at the same stage → · Same stage in other industries →
- Veeda AI is a Toronto-headquartered company incorporated in June 2026 by CEO Sanja Fidler, who led NVIDIA's Toronto AI lab for eight years, together with former NVIDIA colleagues Zan Gojcic (CTO) and Huan Ling (chief scientist).
- The company confirmed that it has raised USD 90 million to date in a seed round co-led by Radical Ventures and Khosla Ventures in August 2026.
- The company says that for robots, learning by trial and error in the real world is dangerous, expensive and slow, and that it builds virtual environments where robots learn through interaction, using multimodal world models trained on sensor and physical data.
Decisions within a year
- On August 19, 2026 it came out publicly and announced its seed funding. The funds go to hiring research staff and to compute for training generative world models.
- It said it is hiring at all four of its bases: Toronto, Zurich, the Bay Area and Singapore.
- From August to October 2026 it opened postings for robotics, simulation, world model, and ML data, performance and operations roles, plus internships. The robotics roles put learning-based robot policies onto real hardware and handle experiments and benchmarks.
People they will look for
- Robot experimentation and data collection engineer — building robot systems for world model evaluation and real hardware experiments
- World model and simulation researcher — research on generative models that create physically realistic virtual environments
- ML data and performance engineer — large-scale sensor data pipelines and training efficiency
For hiring managers
PhD-level talent who has worked on both robot policy learning and generative models is sought by robot and AI labs at large Korean companies and by the Korean bases of global AI companies alike. Veeda has no base in Korea and its posted work locations are Toronto, Zurich, Seattle and California, so these are openings for Korean researchers who want to work abroad.
For job seekers
Experience putting learning-based robot policies onto real robots and validating them in reproducible experiments, or research experience in 3D and simulation generative models, is worth the most at this company right now.
An analysis written from public sources. Sources: BetaKit 2026-08-19 (founding, co-founders, USD 90 million confirmed by the company), Canadian Lawyer 2026-08-19 (Radical and Khosla co-lead, incorporated June 2026, use of funds), Veeda AI website (mission, hiring at four bases), Veeda AI careers page (Ashby, checked 2026-10-26: roles and work locations).
Open jobs
Banseog checks official company careers pages every day. A job is marked closed only when the official page returns a removal (404 or 410).