# Maxinsights

> Markdown twin of https://roboterminal.app/companies/maxinsights · https://roboterminal.app/companies/maxinsights.md

## Pulse

- [@CyberRobooo on X](https://x.com/CyberRobooo/status/2113306432712102371)
  I’ve been watching Physical AI data scale from tens of thousands of hours to millions. But seeing that Maxinsights has collected 2 million hours of people doing chores in real homes genuinely surprise…
- [@techniahqrobot on X](https://x.com/techniahqrobot/status/2113305679880491323)
  Let me say this again: how can a robot learn to grip an object if it can’t sense how much force it’s applying? A camera can see a hand closing around a glass, but it can’t directly measure grip force …
- [@humanoidsdaily on X](https://x.com/humanoidsdaily/status/2113301386997731574)
  Maxinsights raised a $30M seed and launched MaxGlove, a capture glove that records grip force alongside video. Most manipulation datasets are camera-only, so they can't tell a firm grip from a loose o…
- [@techniahqrobot on X](https://x.com/techniahqrobot/status/2111179298052722751)
  Love seeing this level of annotation. Environment to specific hand instruction plus accurate pose and contact all in one. This is what makes VLA training actually work better. Solid work. https://t.co…
- [@CyberRobooo on X](https://x.com/CyberRobooo/status/2102612616355209550)
  A milestone-> Dyna-2 showed that 1M+ hours of human egocentric video can scale robot learning. As the dataset grew from 1K to 1M hours, performance improved on both human and robot tasks, despite usin…

---

Source: Roboterminal · https://roboterminal.app/companies/maxinsights.md