Podcasts + Vodcasts
10.6.2026
Co-founder and CEO Hardik Kabaria sat down with NextGenInfra.io following Vinci’s $250 million Series B at a $1.5 billion valuation. The interview covers what Continuous Physics Reasoning means for hardware engineering, the customer requirements that shaped Vinci’s product, why thermal analysis matters for AI infrastructure, and where the company goes next.
Watch the Interview
Kabaria starts with the problem behind Continuous Physics Reasoning. Every hardware product, from high-bandwidth memory and GPUs to phones, data center racks and satellites, is governed by the same universal laws of physics. The better engineers understand where hot spots will form and where mechanical stresses could cause reliability or manufacturing problems, the better they can design. Kabaria calls physics the compilation layer of hardware. Vinci’s enterprise software makes thermal physics accessible to everyone in a hardware engineering organization, from early concept through hand-off to manufacturing partners.
He describes Vinci as a product company before a technology company, and says the platform was built backward from customer requirements. Customers want their design files kept private and behind their firewall. They want accuracy and high fidelity, not just speed. They want deterministic answers, where the same question returns the same result. And they want a product that works the moment it’s deployed. Vinci’s answer is a model pre-trained on physics and wrapped in an enterprise product that runs behind the customer’s firewall. It requires no fine-tuning and delivers results orders of magnitude faster, at the accuracy engineers already rely on and, in some cases, higher fidelity. Because Vinci trains on physics rather than anyone’s data, every customer gets the same product.
For AI infrastructure, Kabaria explains how one model spans every level of the stack: the back end of the line in 3D integrated circuits, chip packages, PCBs, and full mechanical chassis with heat sinks. Energy balance, stress balance and momentum balance hold at every scale. That lets a model trained on the underlying equations generalize to whatever products customers are building. Vinci has proven this with leading hardware companies in the US, Europe and Asia and is attached to flagship commercial programs. Next, it will expand into fluid dynamics and system-level cooling, electromagnetic signal integrity, power integrity and power delivery networks, and phenomena like plasticity and viscoelasticity. Customers are guiding that expansion directly.
"Physics is owned by nobody. Physics is universal."
Hardik Kabaria
Co-Founder and CEO, Vinci
The requirements Kabaria lists map directly to how the platform is built. Deployment behind the firewall keeps sensitive designs in the customer’s hands. Solver-grounded, deterministic output gives engineers results they can repeat, compare and carry into sign-off. A pre-trained foundation model that needs no fine-tuning means teams can start automating workflows as soon as the product is installed.
Kabaria’s roadmap reflects the longer-term vision of a single physics engine that engineers use across the entire design cycle. Every customer runs the same product, so each new modality Vinci adds, whether fluid dynamics, electromagnetics or material behavior, becomes available across every deployment. Teams can evaluate their designs holistically instead of stitching together separate tools for each type of physics.
Vinci is a frontier lab building the foundation model for the physical world. Its deterministic, solver-grounded systems make physics continuously computable inside production engineering workflows and are already running on flagship programs, shifting physics from an episodic simulation bottleneck to continuous infrastructure for design, manufacturing, and reliability decisions.
