Lyte's LyteVision gives robots unified vision through 4D sensing, RGB imaging, and motion awareness, built by the engineers behind Apple's Face ID and Microsoft Kinect.
Photo source:
Lyte
Robots today have a strange weakness. They can
lift, move, and work tirelessly, yet most of them struggle with something
humans do effortlessly: seeing and understanding the world around them. To give
a robot vision, engineering teams currently bolt together cameras, depth
sensors, and motion units from different suppliers, then spend months writing
software to make them agree with each other. According to McKinsey, over 60
percent of industrial companies lack the internal capability to implement
robotic automation, and sensor integration is a major reason why. The machines
exist, but the eyes remain the hard part.
Lyte, a company headquartered in Mountain View,
California, emerged from stealth in January 2026 to solve exactly this. Its
founders know the territory better than almost anyone. Alexander Shpunt, Arman
Hajati, and Yuval Gerson were key architects behind Apple's depth-sensing
technology, and Shpunt co-founded PrimeSense, the 3D sensing company that
powered Microsoft Kinect before becoming the foundation of Apple's Face ID
platform. In short, the people who taught your phone to recognize your face are
now teaching robots to see.
The company's core product is LyteVision, and
its idea is disarmingly simple. Instead of many separate sensors, it integrates
everything a robot needs to perceive into a single platform. The system
combines 4D sensing, which captures space and movement over time, with RGB
imaging for visual detail and motion awareness for understanding change. All of
it flows to the robot as unified spatial and visual data through one
connection. For engineering teams, this replaces months of integration work
with a single component that already works as a whole.
Under the surface, the platform runs on custom
silicon paired with integrated sensors and software, engineered for safety,
reliability, and performance from the hardware up. Moreover, LyteVision
includes an AI-driven operating layer designed to keep improving alongside
breakthroughs in vision, language, and action models. In practice, that means a
robot's perception can grow more intelligent over time rather than staying
frozen at the moment of purchase.
Versatility is central to the design.
LyteVision supports an unusually wide range of machines, including autonomous
mobile robots in warehouses, robotic arms on production lines, four-legged
quadrupeds, robotaxis, and humanoid robots. A logistics company, for example,
could equip its fleet with the same perception platform its supplier uses on
robotic arms, simplifying everything from training to maintenance. Because the
sensing stack is unified, what a robot learns about seeing the world transfers
across very different bodies.
The industry has taken notice quickly. Lyte
introduced its platform at CES 2026 in Las Vegas, where it earned Best of
Innovation recognition in robotics. For a company presenting its technology
publicly for the first time, recognition at the world's largest technology
showcase placed its approach firmly on the industry's radar.
The timing matters as much as the technology.
The AI robotics market is projected to reach 125 billion dollars by 2030,
driven by labor shortages and mounting pressure on manufacturers to automate.
However, the bottleneck has never been ambition; it has been capability. Every
robot that works safely around people must first perceive people, obstacles,
and movement with complete reliability. Therefore, a dependable perception
layer is not a convenience but the foundation that determines how fast automation
can safely spread into warehouses, hospitals, streets, and eventually homes.
There are fair questions ahead. Lyte is a young
company; its platform has just reached the market, and winning over robot
manufacturers takes time in an industry where trust is earned through
deployment. Even so, the ingredients are unusual: a founding team with a proven
record of shipping perception technology to billions of devices and a product
aimed at the industry's most persistent gap. If robots are going to work
alongside humans at scale, they will need eyes they can rely on, and Lyte has
built its entire company around providing them.
Please subscribe to have unlimited access to our innovations.