Maxinsights launches MaxLoop and MaxGlove, adding live quality control and touch data to physical AI data stack

Maxinsights, a California-based company that says it is building the data layer for physical AI, launched two new products to enhance its data collection stack.

The company announced the launch of MaxLoop, an agent-in-the-loop system that scores and corrects data collection while it happens, and MaxGlove, a sensor glove for capturing tactile touch and force for a new era of data collection.

MaxLoop: Scoring every clip as it arrives

Maxinsights said that real-world footage breaks most annotation tools. Hands disappear behind objects, several people's hands enter the same frame, operators wear gloves and rings, motion blurs.

The company said that it trained its own hand-tracking models on footage it collected itself, and built MaxLoop to score every clip as it arrives. Maxinsights said that operators now get feedback the same day instead of waiting one to two weeks.

Maxinsights said that human auditors are down 90%, from roughly 2,000 to 3,000 to 200, and the company processes 450,000 hours a month. The same models also pull more out of each clip: pose, object states, contact and force. An hour of video that used to teach a model one thing can now teach it several.

MaxGlove: Capturing what a camera can’t see

Next, the company said that MaxGlove, in development with Genesis AI, captures touch and force: how hard a hand grips, how much an object weighs, how contact shifts through a task.

Maxinsights said that foundation models need force data because vision alone can’t tell a robot how an interaction feels. Force and torque signals reveal contact, pressure, resistance and slip, helping robots know whether they’ve grasped an object securely, how much force to apply, and when to adjust.

Maxinsights said that no company has yet to demonstrate a truly scalable way to collect force data: existing setups require lengthy or repeated calibration, while Maxinsights said that its device integrates with egocentric systems and can be deployed reliably across a wide range of environments.

The company said that it manufactures its own because gloves on the market need long calibration cycles and yield only 10 to 20 minutes of usable capture a day. Maxinsights said that MaxGlove goes across the existing collection network, and the company is targeting 1 million hours of force data in 2027.

“We built the whole pipeline ourselves, from the hand-tracking models to the platform that scores every operator's footage as it arrives, because nobody was going to sell us the pieces that make messy real-world data usable,” said Yuguang (YG) Yuan, founder and CEO of Maxinsights. “Robots get deployed into real kitchens, not staged ones, and something unexpected happens every few minutes. Edge cases are the job, and the only way to get them is to record real work in real homes. Collecting one great hour of data is easy, and holding quality across millions of hours and thousands of people is what we built the company to do.”

A full data stack in one place

Maxinsights said that while competitors stage tasks in mock environments and ask operators to slow their hands so the footage is easier to process, the company records real work at natural speed in real homes, restaurants and hotels, so models keep seeing new settings and unplanned moments. Every step from capture to delivery happens in-house.

Maxinsights said that it operates in five steps:

• Collect: First-person (egocentric) footage of real people working, captured in the wild by our own global network across varied places, objects and tasks.
• Check: Automated and human review of every clip, capturing quality, task completion, visibility, sync, metadata.
• Annotate: Proprietary models label each clip across four layers (environment, task, subtask, instruction) plus hand pose, body motion and object interactions.
• Curate: Search and rebalance the corpus by task, scene, geography, behavior, complexity or whatever a client needs.
• Deliver: Ship raw or processed datasets, then feed client and model feedback into the next collection cycle.

The company said that the captured footage comes from people getting paid to record work they were already doing. Maxinsights said that over 5,000 contractors in six countries wear its proprietary camera hardware while they load the dishwasher, make a smoothie or wash a car, capturing clear footage compared to shaky iPhone or GoPro videos.

Maxinsights touts customer validation

The company said that it has captured more than 2 million hours, with 1.5 million hours model-ready and available for immediate delivery. Maxinsights said that it serves over 30 customers, including frontier AI labs and foundation model companies, with eight on recurring contracts.

Specifically, Maxinsights said that two of the most closely watched robotics foundation models of 2026 were trained in part on its data:

• Genesis AI: All 200,000 hours of egocentric data used to train GENE-26.5 were provided exclusively by Maxinsights. The two companies are now developing MaxGlove together to add touch and force data.
• Dyna Robotics: Maxinsights is the largest data provider behind Dyna-2, Dyna's World-Action Model.

“Robots need to learn from the complexity and variation of everyday life, and carefully controlled datasets alone cannot capture that,” said Johnson Wang, cofounder of Genesis AI. “With Genesis 26.5, we focused on footage of people performing everyday tasks naturally, with much of that data provided by Maxinsights, to help our models learn from a broader range of real-world situations”

Maxinsights secures $30 million in seed funding 

In a LinkedIn post, Maxinsights announced $30M in seed funding led by Peak XV Partners.

Yuan previously co-founded Mega Gear Robotics, an industrial robotics company that was acquired.

Kun Wang, co-founder and CTO of Maxinsights, was an early engineer at Google X/Waymo, then co-founded QCraft, which deployed autonomous-driving technology in over 1 million vehicles.

Maxinsights said that the company’s broader team includes engineers from Google, Microsoft, Meta, Waymo and Hulu.