The Robot Data Gap | Physical AI Happy Hour
Join for a casual happy hour with researchers and builders tackling one of physical AI’s hardest questions: What data do robots need next?
We’re bringing together people training robot policies, building collection pipelines, and working with embodied AI to trade perspectives on collecting useful training data at scale. From teleoperated demonstrations and egocentric human video to synthetic data, we’ll explore how teams choose what to collect, measure data quality, and uncover the gaps behind real-world failures.
No pitches, no slides. Just conversations over drinks about collection bottlenecks, task and environment diversity, and what it takes to turn more data into better robot performance.
Come share what you’re learning, compare notes with others facing similar challenges, and meet the people building the next generation of physical AI.
About the Hosts
Frank Tanner | VP, Computer Vision & Robotics, Innodata
Frank leads Innodata's Robotics & Physical AI Lab, advancing the data infrastructure needed to train, evaluate, and validate next-generation robotic systems. His work spans computer vision, physical interaction data, and the challenges of scaling AI from research into real-world applications.
Rishab Mohan | AI GTM, Innodata
Rishab works at the intersection of AI research, emerging technology, and the people building what's next. At Innodata, he develops strategic initiatives across physical AI, robotics, and frontier AI, while curating intimate dinners and gatherings that bring together leading researchers, founders, and executives for meaningful conversations and connections.