JuliaHub has launched Dyad 3.0, its agentic AI platform for hardware engineering, securing a $65 million Series B round led by Dorilton Capital. The investment will accelerate the platform's mission to revolutionize how complex physical systems are designed. Dyad empowers engineering teams to reduce development cycles from months to days, marking a significant leap in physical AI.
A New Era for Physical Engineering
The physical engineering sector has lagged in the AI revolution, constrained by legacy tools. This has created a bottleneck for critical infrastructure projects requiring trillions in investment to meet future needs. JuliaHub's Dyad platform emerges as a modern solution designed to bring the speed and efficiency of AI to hardware innovation.
Dyad 3.0 provides an AI-first environment for modeling and validating industrial systems, acting as an AI assistant for the physical world. It connects autonomous agents with scalable physics simulations and safety analysis to streamline development. CEO Viral Shah describes it as "agentic engineering at scale," generating complete system designs from a single specification.
Strategic Investment and Industry Vision
The $65 million funding round was led by Dorilton Capital, with participation from General Catalyst, AE Ventures, and Bob Muglia. Daniel Freeman of Dorilton Capital highlighted systems modeling as a vital layer where physics, control logic, and AI converge. He praised Dyad as a platform that compiles systems, taking engineers from concept to production in a single environment.
Investors believe JuliaHub is positioned to become a defining company in the emerging Physical AI field. The platform's unique ability to integrate disparate engineering processes into a single, unified environment is a key differentiator. This approach promises to unlock significant productivity gains and accelerate the development of next-generation intelligent systems.
Bridging AI and Physical Reality
Dyad's core strength is its foundation in Scientific Machine Learning (SciML), ensuring AI models adhere to physical laws. Unlike general-purpose AI, which can produce impossible results, Dyad's agents reason based on scientific principles. This is crucial for hardware engineering, where errors can lead to catastrophic failures and significant safety risks.
The platform's cloud-based agents continuously scan scientific knowledge to improve models, while real-world data enhances its digital twins. This allows models to evolve, providing engineers with accurate simulations for testing millions of designs. Engineers remain the essential human-in-the-loop, using Dyad's tools to verify processes and ensure product safety.
Industry Adoption and Partnerships
Fortune 100 companies in aerospace, automotive, and utilities are already using Dyad to streamline their R&D. A partnership with Synopsys integrates Dyad with Ansys TwinAI™, enabling high-fidelity digital twins that combine physics-based simulation with data-driven models. This integration significantly accelerates the digital engineering lifecycle for software-defined systems, according to Synopsys.
In an application with water management company Binnies, a Dyad-powered digital twin predicts pump faults with over 90% accuracy. David Joyce, former CEO of GE Aviation, praised Dyad for operating where physics meets analytics, creating value for customers. These endorsements underscore the platform's potential to move industries from reactive to predictive operations.
With its Dyad 3.0 launch and new investment, JuliaHub is poised to redefine hardware engineering. Its agentic AI, grounded in scientific principles, offers a powerful new paradigm for designing the complex physical systems that underpin modern infrastructure. As industries turn to AI for innovation, Dyad provides a critical tool to accelerate progress safely and efficiently.