FrodoBots built a wearable called RoboCap and open-sourced it. Ten-plus hours of continuous egocentric video capture, designed specifically for real-world data collection outside of controlled lab environments. This piece of hardware barely gets mentioned in any @BitRobotNetwork coverage, but it sits at the intersection of two things the network needs simultaneously: scale and diversity. SN/05 SeeSaw runs on iPhones. That's the mass participation layer, low friction, anyone can contribute, high volume. RoboCap is the precision layer. A wearable purpose-built for egocentric capture produces consistent framing, consistent field of view, consistent sensor data across contributors. When you're training policies for robots that need to understand how humans interact with objects from a first-person perspective, consistency in the capture hardware is not a minor detail. It's the difference between a dataset that trains clean models and one that introduces hardware-level noise into every sample. The open-source decision is also worth noting. FrodoBots didn't productize RoboCap. They published it. Any research team, any lab, any hardware partner can now build on that design, modify it, deploy it in their own data collection programs. If that catches traction in academic robotics, BitRobot ends up with a distributed hardware standard for egocentric capture that feeds back into the network's own training pipelines. That feedback loop is elegant on paper. How many labs actually build on an open-source wearable from a crypto robotics startup is a different kind of test.