NcodiN Secures €16M Seed Round to Overcome AI’s Copper Wall with World’s Smallest Laser

In a move set to redefine AI hardware, Paris-based deep-tech startup NcodiN has raised €16 million in an oversubscribed seed round, positioning itself for industrial-scale production of its pioneering photonic interposer technology.​​

Led by MIG Capital and joined by other global venture capitals – Maverick Silicon, PhotonVentures, Verve Ventures, and returning supporters Elaia, Earlybird, and OVNI, the funding will fast-track NcodiN’s transition from R&D to manufacturing. Proceeds will back new engineering hires, the industrialization of its technology on 300 mm CMOS wafers, a strengthened supply chain, and expanded customer partnerships. The company is also set to establish a Silicon Valley presence while scaling its R&D team for mass manufacturing partnerships.​

Founded in 2023 as a CNRS spin-off, NcodiN is charting a new course in optical interconnects for next-gen computing infrastructure. By embedding the world’s smallest lasers onto silicon, the company promises a scalable, sustainable path forward for AI hardware—poised to power tomorrow’s computational demands far beyond what copper interconnects can offer.​

Pioneering Photonics to Break the Copper Bottleneck

NcodiN’s breakthrough, NConnect, is designed to break through the ‘copper wall’ – the performance and energy ceilings set by conventional electrical interconnects limiting today’s AI systems. At its core is the world’s smallest laser, enabling a densely integrated optical interposer that allows for supercomputer-level capability within a single processor, all while working seamlessly with existing architectures.​

Speaking of the development, Francesco Manegatti, CEO and Co-founder, said, “This funding marks a pivotal milestone for NcodiN. We are delivering the missing piece for the industry’s most pressing challenge: enabling extremely high memory bandwidth to power the AI factories of tomorrow.”

“Our technology unlocks wafer-scale superchips by providing the most energy-efficient interconnects for networking across tens of chiplets – an essential component in the architectures everyone is chasing. As new generations of GPUs and AI accelerators emerge to keep pace with rapidly evolving GenAI algorithms, NcodiN is laying the photonic foundation that makes them possible,” Manegatti added.​​

Broad Recognition and Strong Endorsements

Over the past 18 months, NcodiN has drawn industry recognition for addressing AI’s toughest bottlenecks, demonstrating record-low energy (below 0.1 pJ/bit) prototype nanolasers, integrated detectors, and complete optical links on silicon. The company has also built a cleanroom for rapid prototyping and joint development with industry partners.​​

Memory bandwidth remains a defining challenge in AI. Josh Miner of Maverick Silicon remarked, “Memory bandwidth has become a defining bottleneck in AI, with copper interconnects struggling to deliver the reach and efficiency required for next-generation systems. NcodiN’s photonic interposers unlock memory bandwidth and capacity beyond copper’s limits. We are excited to support NcodiN as they enable architectures that will define the future of AI hardware.””

NcodiN’s strategic advisory board features key technology leaders. Renowned physicist and photonic crystals trailblazer Eli Yablonovitch commented, “NcodiN is addressing the next big opportunity in Silicon Photonics, the Silicon Photonic Interposer, which permits high-bandwidth communication in a multi-chip system, particularly for AI and Machine Learning. To do this, they embed high-performance nanolasers in silicon, thereby combining physics elegance with manufacturability. This has the likelihood of revolutionizing all future large scale cyber systems.​​”

ⓘ As part of our ongoing support for startups and SMEs, LAFFAZ Media publishes feature and resource articles that may include references and links to external websites. These inclusions are selected at our editorial discretion to provide valuable information to our readers. LAFFAZ Media does not control, endorse, or assume responsibility for the content or practices of external websites. For more details, please refer to our Terms and Conditions.