Skip to content Accessibility statement
YorRobots
HomeYorRobotsNews and eventsEventsSeminar series2026Neuracore: The Robot Infrastructure to Accelerate Academic Research

Neuracore: The Robot Infrastructure to Accelerate Academic Research

Seminar

This event has now finished.

Event date
Wednesday 4 March 2026, 3pm to 4pm
Location
In-person and online
ISA 135, Institute for Safe Autonomy (Map)
Admission
Free admission, booking not required

Event details

Robotic research is undergoing a paradigm shift from model-based methods to data-driven approaches, accelerated by recent breakthroughs such as RT-2, Action Chunking Transformers, and the Physical Intelligence (PI)'s work. But collecting high-quality robot data, training at scale, and deploying policies reliably still requires significant infrastructure work—we understand that infrastructure is a means to an end, and that the academic rather focus on research than on building and maintaining it.".

In this talk, I’ll introduce Neuracore, a cloud-based robot learning infrastructure designed to close that gap and help academic labs focus on ideas and iterate much faster. I’ll share the design philosophy behind Neuracore and demonstrate core capabilities: robot data collection, data visualization, cloud training, and deployment to robots.

Neuracore also open-sources its core library. Based on my experience as a core contributor to Neuracore and LeRobot (the widely used open-source robot learning library), I’ll discuss practical strategies for how academic labs can contribute to open-source projects—and how open-source can amplify research impact, reproducibility, and collaboration.

Online Zoom Link: https://york-ac-uk.zoom.us/j/94779725321?pwd=Rf1D0OSDOL1cko6syQ1ChxLYUJ9NYj.1&jst=2

About the speaker

Dr. Ke Wang

Dr. Ke Wang is a Robot Learning Engineer at Neuracore, specializing in developing infrastructure for vision-language-action (VLA) models and reinforcement learning. Previously, he was a Tech Lead at Leap AI, where he led robot learning initiatives, and a Senior Robot Research Engineer at Dyson, focusing on deep reinforcement learning for robotics.

He earned his PhD in Legged Robotics from Imperial College London. His technical interests include reinforcement learning for locomotion and manipulation, as well as developing general and efficient VLA models. Dr. Wang is also an active open-source contributor and a core contributor to the reinforcement learning stack of LeRobot, a widely used open-source robot learning library in the Hugging Face ecosystem.

Venue details

Wheelchair accessible