BLO LAB (Blomega) builds wearable sensing hardware, the GX-1 glove and HC-1 head camera, that captures human hand skill (finger bend, fingertip force, wrist motion, and time-synced POV video) and turns it into training data for robot manipulation and embodied-AI models.

What does the GX-1 glove measure?

The GX-1 records 10 finger-bend channels (2 joints per finger), fingertip grip force on thumb, index and middle fingers, and a 6-axis wrist IMU, all at 100 Hz, hardware-synchronized and video-sync capable.

How can I get paid to capture data for BLO LAB?

Skilled people join Talika (talika.ai), wear the gear, and complete real-world tasks. Every verified hour of skilled work becomes training data and a paycheck; gear ships to your door.

All articles

Education·9 min read·July 7, 2026

In-Situ Capture: Data Where the Work Actually Happens

The kitchen, the workshop, the lab bench, these are the places skilled work happens. A capture stack that has to be shipped to them beats one that expects them to come to it.

There is a specific class of skill that a teleop warehouse can never capture cleanly, and it is a large class. Any skill whose signal depends on the real environment, the actual pot on the actual stove, the real fabric under real light, the actual grease and grit of a working shop, degrades when you move it into a cell. A cook in a lab cooks differently. A seamstress on a warehouse table sews differently. The skill has been decontextualized before the sensor ever fires.

Why context matters more than people expect

Skilled work is not just the hand motion. It is the hand motion in response to the specific environmental cues the worker has spent years learning to read. The sound of a pan starting to smoke. The feel of dough that has been resting for the right amount of time. The tension in a wire that is about to bind. Remove the environment and you remove the cues; keep the motion and you have captured a shadow of the skill.

For contact-rich, long-horizon work, exactly the tasks that foundation models most need, this is not a small effect. It is the difference between demonstrations that transfer to real environments and demonstrations that look right on video and fail in deployment.

In-situ is only possible if the rig goes to the operator

In-situ capture requires wearable hardware that fits in a bag, wireless, battery-powered, and robust enough to survive real work environments. This is exactly the design brief for the BLO LAB suit, glove, and strap set. A rig that has to be shipped to a fixed cell forfeits the entire category.

This is not a knock on teleop; it is a scoping statement. Some tasks belong in cells. Most skilled human work does not. A data program that only captures the tasks that fit into a cell is systematically under-sampling the space of work robots will eventually need to do.

Work with us