Mosaic SoC

Spatial computing — the ability of devices to understand and interact with the three-dimensional world around them — is moving quickly from science fiction toward mainstream hardware. AR glasses, next-generation smartphones, wearables, and IoT sensors all need to perceive depth, map environments, and recognize objects in real time. The problem is that doing this well has historically required the kind of processing power and energy consumption that belongs in a data center, not a pair of glasses that needs to last a full day on a single charge. Mosaic SoC, a Swiss semiconductor startup founded in 2024, is building the chip that makes real-time spatial intelligence possible on low-power, battery-constrained devices. The company's dedicated perception processor handles the heavy lifting — depth sensing, simultaneous localization and mapping, object recognition — with energy efficiency that general-purpose processors simply cannot match. Where a GPU or a high-end mobile processor might accomplish the same tasks at significant power cost, Mosaic SoC's chip is designed specifically for this workload, trading generality for efficiency and delivering results that make always-on spatial awareness viable outside of tethered or plugged-in devices. The target markets are well-defined: AR glasses manufacturers who need perception that does not drain a battery in an hour, smartphone makers looking to add spatial features without thermal or power penalties, and IoT and wearable device makers for whom power budgets are a hard constraint. In all of these cases, the bottleneck has been hardware — not the algorithms, not the software, but the physical chip that has to execute those algorithms within a tight energy envelope. With €3.2 million in pre-seed funding led by Founderful and supported by the Kick Foundation, Mosaic SoC is moving toward its first design-in partnerships with hardware makers. The semiconductor path from design to volume production is long and capital-intensive, but the demand signal from the spatial computing market is strong and growing. The company is at the beginning of that journey, building the foundational hardware layer that could power the next generation of spatially aware devices.