Axiomatic AI, a company building AI infrastructure for science and engineering, has raised $18m in seed funding, according to a news release reported by Built In on 9 March 2026. Engine Ventures led the round, with participation from Kleiner Perkins, Big Sur Ventures and Global Vision Capital, among others.
Pulse 2.0 listed Propagator Ventures and Liquid 2 among the other participants and said the round brings Axiomatic AI's total funding to $25m. No valuation was reported in the sources reviewed.
“AI that cannot justify its reasoning to the level needed for engineering cannot scale into high-stakes technical domains.”Jake Taylor, CEO, Axiomatic AI
What Axiomatic AI does
The company is developing a platform called Axiomatic Intelligence, which it abbreviates as AxI. Built In described it as combining frontier AI models with mathematical and physics-based verification methods, uncertainty quantification and domain expertise. The aim is to help scientists and engineers automate complex workflows while keeping outputs consistent with real-world physical constraints.
Pulse 2.0 explained the problem the company says it is addressing: generative models can suggest designs or solutions, but they typically cannot prove that the outputs comply with the laws of physics, validate their assumptions or quantify uncertainty. Axiomatic AI says it checks correctness across several layers, including physical principles, engineering design constraints and logical reasoning.
The company's website describes a web-based product called Lemma, which combines AI with the tools and workflows scientists and engineers already use, and says its enterprise suite was built first for photonics.
Founders and customers
Pulse 2.0 reported that Axiomatic AI was founded in 2024 by researchers and engineers specialising in physics, photonics and electrical engineering. Named founding team members are Dirk Englund, a professor of electrical engineering and computer science at MIT; Frank Koppens, a group leader and professor at the Institute of Photonic Sciences; Joyce Poon, a professor at the University of Toronto; and Marin Soljacic, a physics professor at MIT.
The chief executive is Jake Taylor, who previously served as assistant director for quantum information science at the White House and as senior adviser for critical and emerging technologies at the National Institute of Standards and Technology.
According to the company, its early-access programme already includes several Fortune 100 and Fortune 500 enterprises, including semiconductor equipment makers, foundries, fabless chip designers, photonics firms and research institutions. It did not name them.
What the money is for
The company said the capital will support broader enterprise deployments and deeper integration of its verification platform into scientific and engineering workflows. Its stated long-term goal is to become the foundational workflow platform for engineering and science organisations working in critical industries.
Israel Ruiz, president and general partner at Engine Ventures, said in the release, as reported by Pulse 2.0, that science and engineering are the backbone of modern civilisation and that the shift from prediction to provable reasoning will define the next era of AI adoption in critical industries.
Taylor said machine learning systems must move beyond assistance into accountable collaboration as demand grows for the hardware underpinning the economy. Pulse 2.0 cited industry forecasts that the US semiconductor sector alone may need about 160,000 more engineering roles by 2032. The sources did not report headcount or revenue.
Company website: axiomatic-ai.com