**The Shift from Cloud AI to On-Device Inference: How Quadric is Riding the Wave**
As I dove deeper into the world of AI, I came across a startup that’s making waves in the industry – Quadric. This chip-IP startup, founded by veterans of early bitcoin mining, is leading the charge in on-device inference, and it’s paying off in a big way.
Headquartered in San Francisco and with an office in Pune, India, Quadric has already seen significant growth. In 2025, the company raked in $15-20 million in licensing revenue, a major jump from the $4 million in 2024. And, as CEO Veerbhan Kheterpal told TechCrunch, they’re targeting up to $35 million this year as they scale up their royalty-driven on-device AI business.
This success has given the company a significant boost, with a post-money valuation of between $270-300 million, up from around $100 million in their 2022 Series B. Quadric has also attracted new investors, announcing a $30 million Series C round led by ACCELERATE Fund, managed by BEENEXT Capital Management, bringing their total funding to $72 million.
**From Automotive to the Mainstream**
Quadric started in the automotive space, where on-device AI can power real-time features like driver assist. But, as Kheterpal explained, the spread of transformer-based models in 2023 pushed inference into “everything,” creating a sharp business inflection over the past 18 months. As more companies try to run AI domestically rather than relying on the cloud, Quadric is positioning itself to power this shift.
“We were trying to build the same CUDA-like or programmable infrastructure for on-device AI,” Kheterpal said. Unlike Nvidia, Quadric doesn’t make chips itself. Instead, they license programmable AI processor IP, which customers can embed into their own silicon, along with a software stack and toolchain to run models, including vision and voice, on-device.
**Sovereign AI and the Future of AI Infrastructure**
Quadric is now looking beyond traditional industrial deployments and into markets exploring “sovereign AI” methods to reduce reliance on U.S.-based infrastructure. The startup is exploring prospects in India and Malaysia and counts Moglix CEO Rahul Garg as a strategic investor helping shape their India “sovereign” approach.
The push is driven by the rising cost of centralized AI infrastructure and the difficulty many countries face in building hyperscale data centers. Instead, Quadric is exploring “distributed AI” setups where inference runs on laptops or small on-premise servers within offices rather than relying on cloud-based services for every query.
As the World Economic Forum recently pointed out, this shift is part of a broader trend in AI infrastructure, with inference moving closer to users and away from purely centralized architectures. EY has also noted the sovereign AI approach gaining traction, with policymakers and industry groups pushing for domestic AI capabilities.
**The Programmable Method**
Quadric is pitching itself as an alternative to chip distributors like Qualcomm and IP suppliers like Synopsys and Cadence. The startup’s programmable processor IP allows customers to support new AI models through software updates rather than redesigning hardware, giving them a competitive edge in an industry where chip development can take years.
But, Quadric is still early in its buildout, with a handful of signed customers so far and much of its longer-term upside dependent on converting today’s licensing deals into high-volume shipments and recurring royalties.
—
Note: I rewrote the article to make it more conversational and easier to follow. I also made sure to keep the original content and quotes from the TechCrunch article, while adjusting the wording and sentence structure to make it more natural and readable. Let me know if you have any other requests!
