𝗣𝗼𝗽𝘂𝗹𝗮𝗿 𝗼𝗽𝗶𝗻𝗶𝗼𝗻: "𝗪𝗲 𝗻𝗲𝗲𝗱 𝗺𝗼𝗿𝗲 𝗱𝗮𝘁𝗮." 𝗔𝗰𝘁𝘂𝗮𝗹 𝗿𝗲𝗮𝗹𝗶𝘁𝘆: 𝟭𝟬𝟬 𝗴𝗿𝗲𝗮𝘁 𝗱𝗲𝗺𝗼𝘀 > 𝟭𝟬,𝟬𝟬𝟬 𝗺𝗲𝗱𝗶𝗼𝗰𝗿𝗲 𝗼𝗻𝗲𝘀. 𝗧𝗵𝗲 𝗺𝘆𝘁𝗵: More demonstrations always mean better models. 𝗧𝗵𝗲 𝗿𝗲𝗮𝗹𝗶𝘁𝘆: I've seen models trained

𝗜'𝘃𝗲 𝗵𝗲𝗮𝗿𝗱 𝘁𝗵𝗶𝘀 𝗮 𝗹𝗼𝘁 𝗿𝗲𝗰𝗲𝗻𝘁𝗹𝘆: "𝗪𝗲 𝘁𝗿𝗮𝗶𝗻𝗲𝗱 𝗼𝘂𝗿 𝗿𝗼𝗯𝗼𝘁 𝗼𝗻 𝗼𝗻𝗲 𝗼𝗯𝗷𝗲𝗰𝘁 𝗮𝗻𝗱 𝗶𝘁 𝗴𝗲𝗻𝗲𝗿𝗮𝗹𝗶𝘀𝗲𝗱 𝘁𝗼 𝗮 𝗻𝗼𝘃𝗲𝗹 𝗼𝗯𝗷𝗲𝗰𝘁 - 𝘁𝗵𝗲𝘀𝗲 𝗻𝗲𝘄 𝗩𝗟𝗔 𝗺𝗼𝗱𝗲𝗹𝘀 𝗮𝗿𝗲 𝗰𝗿𝗮𝘇𝘆!" Let's talk about what's actually

𝗜 𝗵𝗮𝗱 𝘁𝗼 𝘄𝗮𝗹𝗸 𝗳𝗼𝗿 𝗼𝘃𝗲𝗿 𝗮𝗻 𝗵𝗼𝘂𝗿 𝘁𝗼 𝗴𝗲𝘁 𝘁𝗼 𝘁𝗵𝗲 𝗼𝗳𝗳𝗶𝗰𝗲 𝘁𝗼𝗱𝗮𝘆 𝗯𝗲𝗰𝗮𝘂𝘀𝗲 𝗼𝗳 𝗧𝗙𝗟 𝘀𝘁𝗿𝗶𝗸𝗲𝘀. During my walk it made me think about how Japan is building a more sustainable future for both railways and railway workers.

Stop Logging. Start Recording. Every robot interaction is a million-dollar training sample. This robot didn’t just complete a task, it generated intelligence for the future. Traditional logging captures what went wrong. Neuracore records what could be learned. Ready to see how

In robotics today, I see two common approaches to data collection: 𝟭) 𝗥𝗲𝗰𝗼𝗿𝗱 𝗲𝘃𝗲𝗿𝘆𝘁𝗵𝗶𝗻𝗴 𝗮𝘀𝘆𝗻𝗰𝗵𝗿𝗼𝗻𝗼𝘂𝘀𝗹𝘆 𝗶𝗻𝘁𝗼 𝗠𝗖𝗔𝗣 𝗼𝗿 𝗥𝗢𝗦 𝗯𝗮𝗴𝘀. These formats were designed for robotics debugging and replay, not for data-driven learning. • They

𝗧𝗵𝗲 𝗯𝗶𝗴𝗴𝗲𝘀𝘁 𝗯𝗿𝗲𝗮𝗸𝘁𝗵𝗿𝗼𝘂𝗴𝗵𝘀 𝗶𝗻 𝗥𝗼𝗯𝗼𝘁 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗮𝗿𝗲𝗻'𝘁 𝗰𝗼𝗺𝗶𝗻𝗴 𝗳𝗿𝗼𝗺 𝗻𝗲𝘄 𝗮𝗹𝗴𝗼𝗿𝗶𝘁𝗵𝗺𝘀 𝗮𝗻𝘆𝗺𝗼𝗿𝗲. After over a decade in the field of robot learning, I've noticed something: 𝘵𝘩𝘦 𝘣𝘪𝘨𝘨𝘦𝘴𝘵

𝗪𝗵𝗮𝘁 𝗶𝗳 𝗿𝗼𝗯𝗼𝘁 𝗹𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗶𝗻𝗳𝗿𝗮𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗲 𝘁𝗵𝗮𝘁 𝘁𝗼𝗼𝗸 𝟭𝟰 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝘀 𝘁𝘄𝗼 𝘆𝗲𝗮𝗿𝘀 𝘁𝗼 𝗯𝘂𝗶𝗹𝗱... 𝗶𝗻𝘀𝘁𝗲𝗮𝗱 𝘁𝗼𝗼𝗸 𝗱𝗮𝘆𝘀? 1 year ago, I was leading research at the Dyson Robot Learning Lab in London, working with some

Outgrowing our space, not our ambition. Our AgileX Robotics arm and the team have been putting in serious work building Robot Learning Infrastructure for the future. New office incoming. 📍➡️📍

𝗣𝗼𝗽𝘂𝗹𝗮𝗿 𝗼𝗽𝗶𝗻𝗶𝗼𝗻: "Just collect more robot data." 𝗕𝗿𝘂𝘁𝗮𝗹 𝗿𝗲𝗮𝗹𝗶𝘁𝘆: We're facing a 100,000-year data gap. While LLMs like ChatGPT are trained on text datasets equivalent to tens of thousands of human lifetimes of reading, the largest robot manipulation

Will we see humanoid robots deployed en mass by 2030? Our founder and CEO Stephen James whose been working in the field since 2016 shares his thoughts. 𝗪𝗵𝗮𝘁'𝘀 𝘆𝗼𝘂𝗿 𝘁𝗮𝗸𝗲? What do you see as the biggest hurdles for humanoid adoption in your industry?

𝗘𝘃𝗲𝗿𝘆𝗼𝗻𝗲’𝘀 𝘁𝗮𝗹𝗸𝗶𝗻𝗴 𝗮𝗯𝗼𝘂𝘁 “𝗣𝗵𝘆𝘀𝗶𝗰𝗮𝗹 𝗔𝗜" - the idea that we can simulate real-world environments so well that robots trained in simulation will work perfectly in reality. 𝗧𝗵𝗲 𝗽𝗿𝗼𝗺𝗶𝘀𝗲: Train in virtual worlds → deploy anywhere. 𝗧𝗵𝗲

As a newly appointed 𝗔𝘀𝘀𝗶𝘀𝘁𝗮𝗻𝘁 𝗣𝗿𝗼𝗳𝗲𝘀𝘀𝗼𝗿 at Imperial College London, I'm thrilled to announce the 𝗦𝗮𝗳𝗲 𝗪𝗵𝗼𝗹𝗲-𝗯𝗼𝗱𝘆 𝗜𝗻𝘁𝗲𝗹𝗹𝗶𝗴𝗲𝗻𝘁 𝗥𝗼𝗯𝗼𝘁𝗶𝗰𝘀 𝗟𝗮𝗯 (𝗦𝗪𝗜𝗥𝗟) at 𝗜𝗺𝗽𝗲𝗿𝗶𝗮𝗹 𝗖𝗼𝗹𝗹𝗲𝗴𝗲 𝗟𝗼𝗻𝗱𝗼𝗻. 𝗦𝗮𝗳𝗲 𝗪𝗵𝗼𝗹𝗲-𝗯𝗼𝗱𝘆

𝗪𝗵𝗮𝘁 𝗱𝗼𝗲𝘀 𝘁𝗵𝗲 𝗨𝗦/𝗨𝗞 𝗧𝗲𝗰𝗵 𝗣𝗿𝗼𝘀𝗽𝗲𝗿𝗶𝘁𝘆 𝗗𝗲𝗮𝗹 𝗺𝗲𝗮𝗻 𝗳𝗼𝗿 𝘁𝗵𝗲 𝗿𝗼𝗯𝗼𝘁𝗶𝗰𝘀 𝗶𝗻𝗱𝘂𝘀𝘁𝗿𝘆? The US/UK Tech Prosperity Deal signed last week isn't just another diplomatic handshake – it's potentially reshaping how robotics development

𝗗𝗚𝗜𝗦𝗧 𝗿𝗲𝘀𝗲𝗮𝗿𝗰𝗵𝗲𝗿𝘀 𝗵𝗮𝘃𝗲 𝘁𝗮𝘂𝗴𝗵𝘁 𝗿𝗼𝗯𝗼𝘁𝘀 𝘁𝗼 𝗻𝗮𝘃𝗶𝗴𝗮𝘁𝗲 𝗹𝗶𝗸𝗲 𝗵𝘂𝗺𝗮𝗻𝘀 𝗯𝘆 𝘀𝗽𝗿𝗲𝗮𝗱𝗶𝗻𝗴 (𝗮𝗻𝗱 𝗳𝗼𝗿𝗴𝗲𝘁𝘁𝗶𝗻𝗴) 𝗶𝗻𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝗼𝗻, 𝗯𝗼𝗼𝘀𝘁𝗶𝗻𝗴 𝗲𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝗰𝘆 𝗯𝘆 𝟯𝟬%. Their “Physical AI” system lets

"𝗪𝗵𝘆 𝗰𝗮𝗻'𝘁 𝘄𝗲 𝗷𝘂𝘀𝘁 𝘁𝗮𝗸𝗲 𝗚𝗣𝗧-𝟰, 𝗽𝗹𝘂𝗴 𝗶𝘁 𝗶𝗻𝘁𝗼 𝗮 𝗿𝗼𝗯𝗼𝘁, 𝗮𝗻𝗱 𝗰𝗮𝗹𝗹 𝗶𝘁 𝗮 𝗱𝗮𝘆?" Because foundation models for robotics face fundamentally different challenges than language models. Let me explain the technical reality:

The Neuracore team has officially moved into their new office! As we scale our mission to make robotics data collection seamless and accessible, we’re expanding our space to welcome a growing team. Our new robotics development team will be joining us soon and we couldn’t be

𝗧𝗵𝗲 𝗽𝗿𝗲𝗱𝗶𝗰𝘁𝗶𝗼𝗻: Morgan Stanley forecasts a $5 trillion humanoid robot market by 2050. 𝗧𝗵𝗲 𝗰𝗵𝗮𝗹𝗹𝗲𝗻𝗴𝗲: Getting there will be harder than most people realise. Let me share some perspective on what it actually takes: 𝗖𝘂𝗿𝗿𝗲𝗻𝘁 𝗽𝗿𝗼𝗴𝗿𝗲𝘀𝘀 (2025):

𝗡𝗶𝗻𝗲 𝘆𝗲𝗮𝗿𝘀 𝗮𝗴𝗼, 𝗜 𝘀𝘁𝗼𝗼𝗱 𝗶𝗻 𝗮 𝘀𝗺𝗮𝗹𝗹 𝗚𝗼𝗼𝗴𝗹𝗲 𝗰𝗼𝗻𝗳𝗲𝗿𝗲𝗻𝗰𝗲 𝗿𝗼𝗼𝗺 𝘄𝗶𝘁𝗵 𝟭𝟬𝟬 𝗽𝗲𝗼𝗽𝗹𝗲, 𝘀𝗵𝗮𝗿𝗶𝗻𝗴 𝗺𝘆 𝗳𝗶𝗿𝘀𝘁 𝗣𝗵𝗗 𝘄𝗼𝗿𝗸. 𝗧𝗼𝗱𝗮𝘆, 𝘁𝗵𝗮𝘁 𝘀𝗺𝗮𝗹𝗹 𝗿𝗼𝗼𝗺 𝗵𝗮𝘀 𝗯𝗲𝗰𝗼𝗺𝗲 𝗮 $𝟭𝟲.𝟱𝗕 𝗶𝗻𝗱𝘂𝘀𝘁𝗿𝘆...