alqem Raises €8 Million to Advance AI Materials Discovery
  • News
  • Europe

alqem Raises €8 Million to Advance AI Materials Discovery

The deep-tech startup will scale its platform for rare-earth-free magnet discovery

7/6/2026
Ghita Khalfaoui
Back to News

Munich-based deep-tech startup alqem has raised €8 million in pre-seed funding to expand an artificial intelligence-driven platform designed to discover and commercialize advanced materials. The round was co-led by UVC Partners and Union Square Ventures, supporting the company’s initial focus on rare-earth-free permanent magnets. Alqem aims to shorten the process of moving from material prediction to industrial validation, addressing supply-chain constraints affecting sectors such as electric mobility, renewable energy, robotics, and defense.


Building a Materials Discovery Platform

The company is developing its discovery engine around a large database of predicted crystalline materials and proprietary datasets covering material properties. Its platform combines physics-informed AI with laboratory synthesis and characterization, allowing the team to test candidates rather than relying solely on computational predictions. Alqem said this integrated approach is intended to narrow hundreds of millions of potential compounds into smaller groups of materials suitable for experimental evaluation.

Permanent magnets represent the startup’s first target market because they are essential components in electric vehicles, wind turbines, industrial equipment, and defense technologies. Current supply chains for advanced magnets remain heavily concentrated, with China accounting for a significant share of global production. Alqem believes its work on alternatives that do not rely on rare-earth elements could help create more resilient and diversified materials supply chains.

From Open Research to Commercial Development

Alqem was founded in 2026 by Dr. Hanh Nguyen, who serves as chief executive officer, Dr. Tiago Cerqueira, chief technology officer, and Prof. Milan Allan, chief scientific officer. Cerqueira previously co-developed Alexandria, an open materials database that has been used by researchers and companies working on AI-based materials discovery. The startup is now building on that scientific foundation with its proprietary al-mine database, al-oracle training datasets, and in-house experimental capabilities.

The company said its materials database is designed to prioritize compounds that avoid rare-earth, toxic, and high-cost elements. Its AI system is also intended to improve through repeated cycles of computational screening, synthesis, and experimental analysis. According to alqem, this feedback loop could reduce timelines for identifying viable materials from decades to years, or potentially months for specific applications.

Scientific and Industrial Collaboration

Alqem is collaborating with the Max Planck Institute for Chemical Physics of Solids in Dresden on next-generation magnetic materials. The partnership is led by Prof. Claudia Felser, who serves as a scientific advisor to the company alongside Prof. Miguel Marques and former McKinsey senior partner Michael Viertler. Additional research partners include LMU Munich, the Technical University of Munich, Técnico Lisbon, the University of Porto, and the University of Coimbra.

The startup operates across Munich and Coimbra and plans to expand its team as it develops its materials platform. It is also part of the UnternehmerTUM ecosystem, which connects technology companies with researchers, corporate partners, and investors. UVC Partners said the company combines scientific expertise with an opportunity to address a strategic industrial issue for Europe, particularly as demand grows for materials used in clean energy and advanced manufacturing.


Alqem’s €8 million pre-seed round gives the company resources to advance its rare-earth-free magnet pipeline and further develop its AI-enabled discovery engine. While permanent magnets are its first application, the startup intends for its technology to support the identification of materials across multiple industrial categories. Its progress will depend on whether its computational predictions can consistently translate into scalable, commercially relevant materials through laboratory validation and future manufacturing partnerships.