Singapore-based Staera Semiconductor has raised US$4.6 million in pre-seed funding to develop optical interconnect technology designed to address data-transfer bottlenecks in artificial intelligence infrastructure.
The company, formerly known as New Silicon Corporation, emerged from stealth on Nov. 11, 2026, with Lightwire, a technology that integrates LED-based optical links directly into silicon chips to move data between processors and memory.
The funding round was led by Openspace Capital, with participation from SGInnovate, Black Kite Capital, 28 Holdings, In Group and Decimus Technologies.
Staera said conventional copper connections are increasingly constrained by the distances over which they must transfer data in AI systems, particularly between chips, circuit boards and server racks. Its Lightwire technology replaces electrical links with light generated inside the chip, using gallium nitride (GaN) LEDs and silicon nitride waveguides alongside silicon complementary metal-oxide-semiconductor (CMOS) circuits.
Unlike conventional optical interconnects, which typically use separate light sources and components attached to chips, Staera integrates the complete optical link into the silicon wafer. The company said the approach is designed to work with existing semiconductor manufacturing processes and packaging.
Lightwire is intended for use in bridge or interposer chips that connect processors and memory in AI hardware. Staera said the technology is designed to consume less than one-tenth the energy of current co-packaged optical links.
The company estimates that moving 24 terabytes of data per second would require less than 50 watts with Lightwire, compared with nearly 1 kilowatt using co-packaged optics.
“AI’s bottleneck is no longer how fast chips compute, but how fast and far they can move data,” said Dr. Andrew Kim, CEO of Staera.
Kim said the funding would help move Lightwire from research and development toward customer engagement.
Staera said its manufacturing process combines silicon CMOS circuits, GaN LED light sources and silicon nitride waveguides on a single wafer. It reported yields of 98% to 99% from processes used to integrate GaN LEDs into silicon chips at commercial foundries since 2016.
The company lists a data rate of 4 to 8 gigabits per second per link, with more than 600 links per millimeter of chip edge. It also says Lightwire is designed to operate below 0.2 picojoules per bit, compared with 5 picojoules per bit for co-packaged optics.
Other specifications cited by Staera include a reach beyond 50 millimeters, compared with 2 millimeters for UCIe-A die-to-die links, and a Lightwire-based high-bandwidth memory design targeting more than 24 terabytes per second.
The company plans to use the new funding to advance its technology, expand its team in the United States and strengthen engagement with customers and supply chain partners. It is targeting commercial release in 2030.
Hian Goh, co-founder of Openspace Capital, said the technology could help address the growing challenge of moving data between chips without a corresponding increase in power consumption and heat.
Staera originated as a spin-off from the Low Energy Electronic Systems interdisciplinary research group at the Singapore-MIT Alliance for Research and Technology. In May 2026, it won the third ECTC Startup Competition at the IEEE Electronic Components and Technology Conference and placed second in the Lam Capital Venture Competition.
Full disclosure: All news articles published on the TechSabado website are written by human journalists, unless otherwise specified. Final text editing is also performed by human editors, with artificial intelligence (AI) used only to assist with additional grammar and style guide corrections..
————————————————————————-
TEN YEARS OF TECHSABADO!
PLEASE LIKE our FACEBOOK PAGE and SUBSCRIBE to OUR YOUTUBE & SPOTIFY channels.
WATCH TECHSABADO ON OUR YOUTUBE CHANNEL:
WE ARE ALSO ON SPOTIFY
WATCH OUR OTHER YOUTUBE CHANNELS:
PLEASE LIKE our FACEBOOK PAGE and SUBSCRIBE to OUR YOUTUBE CHANNEL.







Leave a Reply