Polaron has secured $8 million to address a long-standing bottleneck in materials manufacturing: the missing intelligence layer that explains how processing affects performance. The London-based company, spun out of Imperial College London after seven years of research, is developing technology that helps engineers understand, design, and optimise advanced materials with unprecedented clarity.
Backed by Racine2, an impact-focused fund led by Serena and Makesense, along with Speedinvest, Futurepresent, and several industrial AI veterans, Polaron now plans to expand its engineering team, accelerate the deployment of its design tools, and meet rising customer demand across automotive, energy, and other manufacturing-heavy sectors.
When asked about Polaron’s valuation, CEO Isaac Squires shared: “We’re not disclosing this.”
A blind spot in modern manufacturing
While factories have automated production for over a century, the understanding of materials themselves remains surprisingly manual. Issac continued: “Manufacturing has been automated for decades, but understanding how materials behave is still manual, fragmented, and slow. Engineers often rely on intuition and trial and error to connect how a material is made with how it performs. Polaron was created to turn that hidden knowledge into systematic, reusable intelligence.”
Grains, pores, phases and defects form a material’s internal architecture, visible under a microscope like a physical fingerprint. These microstructures determine how strong something is, how long it lasts, and how likely it is to fail. Yet connecting these microscopic patterns to performance has historically required laborious analysis.
Turning microstructure into actionable insight
Co-founded by Isaac Squires, Steve Kench and Sam Cooper, Polaron trains advanced AI models on real microscopy images paired with measured material properties. The platform can interpret microstructures, explain performance variations and guide engineers toward better processing decisions, compressing thousands of hours of manual work into minutes.
Its capabilities unlock insights previously out of reach, reconstructing 3D material structures from standard 2D images, rapidly identifying complex features within micrographs, and revealing relationships that were once hidden behind slow experimentation.
Early deployments are already delivering results. Global manufacturers, including electric vehicle makers responsible for over one-third of worldwide EV production, have used Polaron’s tools to design next-generation battery electrodes. In one case, the approach improved energy density by more than 10%, a significant leap in an industry where even single-digit gains matter.
Squires noted a striking example: “In battery manufacturing, changing electrode compression pressure alone can alter fast-charge performance by up to 10×, using the same materials and chemistry. The difference comes from process, not ingredients.”
What about diversity?
Talking about diversity, Issac stated: “We’re currently a team of nine and expect to at least double in size before the end of the year. As we scale, we’re committed to building a diverse, multidisciplinary team.”
What sets Polaron apart from others?
The company does not see itself competing directly with a single category. As Squires continued, “Polaron does not compete directly with a single category. The platform links three things in a single system, including how a material was processed, what its internal structure looks like, and how it performs.”
This allows engineers to explore thousands of process variations digitally rather than through months or years of trial and error.
Generative design for real-world materials
With its characterisation engine in place, Polaron is moving deeper into generative design. By learning the process-structure-property relationships of materials, the system can explore vast design spaces to propose optimal configurations and the exact processing routes needed to produce them at scale.
This bridges the persistent divide between lab discovery and industrial manufacturability. Whether dealing with metals, ceramics, polymers or composites, engineers gain a tool that not only diagnoses material behaviour but also charts the path toward better-performing products.
Future plans
Polaron positions itself as a next-generation materials company, building technology that makes the physical world more understandable and more designable. Its goal is to become the intelligence layer that modern manufacturing has long lacked.
Issac concluded, “Our focus is to expand deployment across new materials and sectors, deepen use inside customer R&D and manufacturing workflows, advance generative design capabilities for materials and shorten development cycles, and improve performance outcomes for customers.”
Isaac Squires, CEO and Co-founder of Polaron, said: “For 150 years, the industry has used machines to shape materials. Now, we are teaching machines to understand them. Polaron is building an intelligence layer powered by the world’s materials data for faster discovery, better design, and a new generation of advanced materials.”
Alix Trébaol, investor at Serena, added: “In materials, AI is commoditising atomistic discovery. The winners will be the ones who can predict real-world industrial manufacturability. No one but Polaron knows how to do this today.”
Florian Obst, Principal investor with Speedinvest’s AI & Infra investment team: “What impressed us about Polaron is its focus on the point where materials innovation often breaks down: translating scientific insight into manufacturable reality. By grounding AI in real microstructural data and industrial constraints, Polaron is building a platform that can accelerate how advanced materials move from research into production.”
