CScale Raises $145M for AI Networking with Nvidia and Intel Backing

CScale Secures $145 Million to Transform AI Networking

CScale, a Palo Alto startup developing optical interconnect for AI data centers, has raised $145 million in Series C funding, bringing its total funding to $188 million1

1

2

2

. Atreides Management led the round, with Valor Equity Partners and Premji Invest serving as co-leads2

2

. Nvidia and Intel joined as the company's first strategic investors, marking a significant endorsement of CScale's approach to AI networking technology2

2

. Existing investors Sutter Hill Ventures and Maverick Silicon also participated in the funding round announced Wednesday1

1

.

Replacing Copper Cables with Fiber-Optic Connections for AI Chips

The company is tackling a fundamental challenge in AI infrastructure by replacing copper cables that connect AI chips inside data center servers with fiber-optic connections. Copper cables, while fast and reliable, have limited reach that forces chip designers like Nvidia and Advanced Micro Devices to pack processors tightly together inside servers1

1

. This tight packing generates excessive heat, requiring elaborate liquid cooling systems1

1

. CScale joins a scramble across Silicon Valley where public companies and startups are pouring billions of dollars into replacing these connections with optical systems that use flashes of laser light instead of electrons to carry data1

1

.

Solving the Laser Failure Problem in AI Optical Interconnect

CScale's core innovation addresses a critical weakness in current optical interconnect approaches. Traditional lasers used in optical systems are prone to frequent failures, forcing many companies to keep lasers separate from chips so they can be replaced1

1

. CScale CEO Martin Lund explained that the company will integrate lasers onto chips in a way that does not interrupt data flow when lasers fail1

1

. "We're designing the interconnect for continuity. Lasers will fail. Compute shouldn't," Lund stated2

2

. Gavin Baker, managing partner and chief investment officer at Atreides, emphasized the stakes: "Every optical failure is a compute failure"2

2

.

Linking Thousands of AI Accelerators at Scale

Optical interconnect becomes essential when a single AI system spans dozens of racks, enabling linking thousands of AI accelerators to work together like one much larger computer2

2

. CScale is building an integrated light engine specifically for this purpose2

2

. While Lund declined to provide specifics about CScale's technology, he noted the company is "not using any exotic technologies" but rather "using proven technologies, integrating them together in a novel way, but preserving the same trusted fiber technology that has been out there and shipping in the industry for many, many years now"1

1

.

Leadership and Timeline for Market Entry

Martin Lund, who leads CScale as CEO, left his position overseeing chip and hardware systems for networking giant Cisco Systems earlier this year1

1

. He previously built Broadcom's switching business and held senior roles at Microsoft and Cadence, most recently running Cisco's Common Hardware Group covering silicon, hardware systems and optics2

2

. Founder and chief technology officer Sanjai Kohli co-founded SiRF, which brought GPS to the mass market, and later founded Inovi, which Facebook acquired in 20142

2

. Founded in 2023, CScale now has about 85 employees and plans to ship its chips by 20281

1

2

2

. The venture capital will fund technology development and market entry2

2

. CScale joins a wave of substantial funding rounds for startups building AI networking infrastructure, including Lumilens with $700 million in August, Celero with $275 million this month, and Eliyan's $145 million Series C in July2

2

.