KalphaTech develops new technologies based on physical models enhanced with artificial intelligence algorithms. We automate expert knowledge and crystallographic analysis with the aim of accelerating research and development, improving quality control, and enhancing production in the pharmaceutical, chemical, and biotechnology industries. Our products extract reliable quantitative and qualitative information—such as the number of components in a solid material, their fingerprints and quantities—or solve the crystal structure of new and unknown phases directly from laboratory X-ray diffraction measurements. The KalphaTech team consists of chemists, physicists, and mathematicians with extensive experience in crystallography, solid-state chemistry, experimental physics, machine learning, and deep learning applied across various domains. Our team has over 20 years of research experience, more than 3 years of industry experience, and its members have published over 120 relevant scientific papers. Some team members are also former startup founders with 4 years of experience in the startup ecosystem, having participated in multiple accelerator programs (Parsec, BlueInvest, ZICER, BlueBioValue).
Crystallography plays a crucial role in materials and solid-state research, especially in understanding the atomic structure that defines most material properties. To decipher the atomic structure, one of the most important and oldest “modern” analytical methods is used—X-ray diffraction on a solid sample, which makes it possible to “look inside” the structure of a crystal. The importance of X-ray diffraction is underscored by the fact that it has been directly or indirectly responsible for more than 30 Nobel Prizes. Besides atomic structure, X-ray diffraction is used for qualitative and quantitative analysis because it allows the precise identification of a material’s “fingerprint”. For this reason, it is not surprising that it is present in nearly every major industry, including pharmaceuticals, chemicals, electronics, construction, nuclear, and energy.
Crystallographic research has led to the development of new materials with improved properties, such as greater strength, better thermal conductivity, or enhanced optical characteristics. Its role is indispensable in discovering new drugs or improving existing ones, and crystallographic data provide a strong advantage in protecting intellectual property for new substances. Therefore, the ability to determine atomic structures more quickly is extremely important in the pharmaceutical and chemical industries.
Problem
Crystallographic data are generated daily in large quantities, but the process takes place in specialized institutions, and their interpretation is costly and requires time and effort from experts—of whom there are few. Often, timely and reliable information obtained from X-ray analysis can save months of work, unnecessary testing, and significant financial resources. Despite the wide availability of this method, its full technological potential in industry and academia remains limited due to specific resource constraints. KalphaTech addresses this problem by offering innovative solutions.
Solution
KalphaTech develops new analytical methods that pair with already widely used laboratory instruments, enabling 10× faster analysis of crystallographic data and more than 50% higher success rates in material identification. This allows faster, cheaper, and more reliable acquisition of information used throughout the daily value chain, whether in manufacturing, quality control, or research. KalphaTech automates expert knowledge and increases competitiveness for all its users.
Team members
The multidisciplinary KalphaTech team consists of five PhDs in chemistry, physics, and mathematics—Ivan Halasz, Tomislav Ivek, Domagoj Vlah, Edi Topić, and Stipe Lukin. After years of scientific collaboration and friendship, they came together to tackle pressing problems in crystallography, leading to the creation of the KalphaTech concept.
Tomislav Ivek is an expert in spectroscopic techniques of crystalline matter under extreme conditions, data analysis, AI and Python. He completed his PhD in physics at the University of Zagreb and spent nearly three years as a postdoctoral researcher at the University of Stuttgart. With 18 years of experience in experimental solid-state and low-temperature physics and 5 years in deep learning, he has published around 40 research articles. At KalphaTech, he works on DevOps procedures and developing and training new ML models.
Domagoj Vlah, an associate professor of mathematics at the University of Zagreb’s Faculty of Electrical Engineering and Computing, has extensive experience in theoretical and numerical mathematics. Over the past five years, he has applied deep learning to three mathematical areas where it was previously unknown: extreme value theory, bilevel optimization, and number theory (elliptic curves). He has won two Data Challenge awards and actively promotes the application of deep learning in mathematics. In KalphaTech, he is responsible for mathematical modelling and development.
Edi Topić recently earned his PhD in chemistry (University of Zagreb). He has more than 3 years of industry and over 5 years of academic experience in solid-state analysis and characterization. His research focuses on crystallography, structural analysis, analytical method development, metrology, data mining, and scientific programming. He is the co-author of 11 papers and four patents. At KalphaTech, he works as a crystallographer and expert in X-ray diffraction modelling.
Stipe Lukin holds a PhD in chemistry and spent nearly 10 years at the Ruđer Bošković Institute. He has received numerous awards, including the 2020 National Science Award. With over 25 scientific publications and extensive industry-related experience through several accelerator programs (Parsec, BlueInvest, BlueBioValue, Climate KIC, ZICER, WIPO), he is also the co-founder and CTO of the startup SeaCras. At KalphaTech, he is responsible for business and development strategy.
Ivan Halasz earned a PhD in solid-state chemistry and has been working at Ruđer Bošković Institute since 2012, where he leads the Laboratory for Solid-State Synthesis and Catalysis. For two decades he has worked with crystallography and X-ray diffraction, including two years of specialization at the Max Planck Institute in Stuttgart. He has published over 100 scientific papers and received several awards, including the National Science Award (2015). At KalphaTech, he focuses on analysing and optimizing ML results.
Motivation for Application and Vision for the End of the Nuqleus Program
As a deep-tech startup, KalphaTech fits perfectly into the Nuqleus program, which offers access to specialized workshops and resources not typically found in academic settings but essential for building a successful company. Nuqleus provides a unique opportunity to transfer knowledge across the team and strengthen the project’s business structure. The program also enables networking with diverse participants, mentors, and investors.
KalphaTech’s vision is to become a global technology startup in crystallography, supplying innovative solutions to leading global scientific and research-equipment companies. Through its offerings, KalphaTech aims to enhance the materials industry and enable clients to fully leverage crystallography in all aspects of their operations.
