Nexthop AI —Building the most efficient Al infrastructure for the world's largest cloud operators

At Nexthop AI, we envision a future where AI infrastructure seamlessly scales to meet the demands of the largest hyperscale cloud environments, unlocking unprecedented operational efficiency and connectivity.

We are committed to redefining the foundations of AI networking by co-developing highly specialized hardware and software solutions that empower the world’s leading cloud operators to handle the increasing complexity and scale of AI workloads.

By leveraging cutting-edge technologies and a collaborative approach, Nexthop AI drives the transformation of AI infrastructure into a streamlined, efficient, and customized ecosystem that shapes the future of cloud computing.

We've been watching the AI infrastructure space heat up, and Nexthop AI caught our attention for all the right reasons. This isn't another chatbot startup — it's a company tackling the unglamorous but critical plumbing that makes AI possible at hyperscale.

What impressed us most? They secured $110 million in Series A funding right out of the gate in 2024. That's not typical seed money — that's "we're serious about competing with the big players" money.

The Pedigree Factor

Founder Anshul Sadana brings nearly two decades of networking expertise from Arista Networks and Cisco Systems. We've seen plenty of AI startups led by brilliant researchers, but Sadana's background in data center networking feels perfectly suited for this moment. He's not learning infrastructure on the fly — he helped build the foundations that today's cloud giants run on.

The team composition tells a similar story. Veterans from Arista, Broadcom, Cisco, NVIDIA, Google, and Amazon don't typically jump ship unless they see something special brewing.

A Smart Market Play

Here's what we find clever about Nexthop's approach: they're not trying to reinvent networking from scratch. Instead, they're using a Joint Development Manufacturer model to create customized solutions with proven merchant silicon chips like Broadcom's. It's pragmatic engineering that gets products to market faster.

The focus on both air-cooled and liquid-cooled options shows they understand the real-world constraints hyperscalers face. Not every data center can handle liquid cooling, and flexibility matters when you're deploying at massive scale.

The Hyperscale Gamble

Nexthop is betting everything on hyperscale cloud providers and AI operators — essentially the world's largest tech companies. It's a high-risk, high-reward strategy that could pay off handsomely if they execute well.

We appreciate their focus, even if it means a smaller total addressable market initially. Building networking infrastructure for Google or Amazon requires a different level of precision than selling to mid-market companies. If they can satisfy hyperscale demands, expanding downmarket later becomes much easier.