Register for the Zoom

Build better computer vision models.

  • Annotate samples
  • Curate datasets
  • Evaluate models
View All Events

Physical AI, ML and Computer Vision Meetup: Robotic Simulation, Infrastructure and Development Tools - March 24, 2027

Mar 24, 2027
9:00 AM - 11:00 AM PDT
Online. Register for the Zoom!
Speakers
About this event
Join us for the Physical AI, ML and Computer Vision Meetup series focused on robotics use cases. In this session we will focus on robotic simulation, infrastructure and development tools.
This layer provides the simulators, cloud infrastructure, and developer tools (the picks and shovels) that make building and operating robots faster and more repeatable, from simulation and fleet ops to config management and deployment.
This virtual series runs every other Wednesday from January 13 to May 5, 2027. Register once and you are registered for every upcoming session in the series.
Schedule
Generating and Evaluating Simulated Robotics Data with Voxel51
Self-healing machine fleets: a case for AI-native operations
Maintaining robotics and machine fleets requires engineers to connect information across manuals, telemetry, maintenance records, and asset relationships. This session introduces Millwright, a modern workspace with a built-in AI diagnostician designed to support that process.
Through a realistic demonstration using mock data, the presentation will walk through AI-guided root cause analysis, showing how the model reasons across manuals and procedures, real-time telemetry, ticket and work order history, and the site’s asset hierarchy to investigate a fault. Attendees will see how these sources inform the diagnostic process and help engineers identify potential causes.
Faster Physical AI Development with CoreWeave
Building robots and autonomous systems means training on huge multimodal datasets, running simulation at scale, and iterating on models quickly. In this short session, CoreWeave's Head of Physical AI shares how teams are building their physical AI infrastructure today, where the bottlenecks are, and what is coming next to make it easier to go from data to deployed models.