Former Meta AR/VR executives raise $21 million to build sensors for the AI era
Elio's founders, Nadav Grossinger and Nitay Romano, spent seven years developing physical sensing systems at Meta. Now they are betting that artificial intelligence needs a fundamentally different way to see the world.
Israeli startup Elio has raised $21 million in Series A funding to commercialize a new type of sensor that its founders believe could fundamentally change how machines perceive the world.
The round was led by Innovation Endeavors and Xora, with participation from Kevin Weil and Scribble VC. Existing investors UpWest and Resolute Ventures also participated after leading the company's previous financing.
Rather than improving image quality for human viewers, Elio is attempting to redesign sensing around the needs of artificial intelligence systems. The company's technology allows AI to determine, in real time, what information a sensor should capture instead of recording a fixed image that software analyzes afterward.
The company was founded by CEO Nadav Grossinger and CTO Nitay Romano, who have collaborated at the intersection of optics and artificial intelligence for more than two decades.
The pair previously co-developed Pebbles Interfaces, which was acquired by Meta, before spending seven years leading development of the physical sensing technologies used in the company's augmented and virtual reality headsets. Earlier in his career, Grossinger founded ColoRight, which was acquired by L'Oréal, while Romano previously served as Chief Optical Scientist at Holo/Or, where he worked on diffractive optics that now underpin Elio's technology.
Today's cameras and optical sensors are largely built around the limitations and strengths of human vision. Elio argues that this approach forces AI systems to process vast amounts of unnecessary visual information before identifying what matters.
Its alternative embeds computation directly into the optical system itself. Dynamic optical layers made of micromirrors process incoming light before it reaches the image sensor, extracting information that conventional optical systems would discard. The company's AI models continuously learn how the optics behave and compensate for them in real time, enabling the system to identify objects and materials through their physical characteristics rather than relying solely on pixels.
The company describes the result as sensors that increasingly behave like software rather than hardware. Instead of becoming obsolete once deployed, the same sensing module can gain new capabilities over time through software improvements, potentially replacing multiple specialized sensors with a single adaptive platform.
Elio believes the approach could have applications across industries where sensing is a critical component.
In life sciences, the technology could enable researchers to observe living cells responding to drugs over time without chemically staining or destroying samples. In semiconductor manufacturing, it could allow engineers to inspect defects hidden beneath stacked layers of chips without cutting wafers apart. For robotics, the company says its platform combines multiple sensing capabilities within a single self-calibrating module that adapts automatically as environments change.
Defense is another potential application. Elio says its technology could improve the detection of small, fast-moving targets such as drones over long distances, including in challenging conditions such as darkness or haze that limit conventional optical systems.
