Matmerize lives at the intersection of materials science and data science and seeks to transform and accelerate industrial materials development at scale. Our mission is to revolutionize and modernize the traditional R&D environment by developing intuitive, data-driven tools for formulation design.
Who We Are
Matmerize is a Software as a Service (SaaS) company committed to changing the field of research and development. We developed our own material informatics platform, PolymRize, backed by years of academic research. PolymRize is a cloud-based software hosted on AWS, accessed through a web-based version or via API.
How We Innovate
Traditional R&D involves significant material and operational costs and can take years to complete. Using PolymRize, custom models can be built, or stock models can be used to predict any number of material properties.
Using the prediction capabilities, candidate lists can be screens for the most promising species for design parameters,. using our unique polymer fingerprinting technique, a dataset of molecular structures and formulations can be turned into a machine learning model. This ML model is then used for property prediction and material design.
Our unique fingerprint represents materials on different levels, from atomic, block, chain, to formulations -level descriptors. Using our screening technology, we can optimize multiple properties at once, outputting materials that meet the desired criteria.
At the current rate of plastic production increase, the ocean will contain more plastic than fish by weight in 2050.
Environmental conservation has been an afterthought of polymer formulation design for far too long. Despite the tremendous utility of petroleum-based plastics in medicine, food packaging, and other industries, these materials are destroying the health of our planet and our population.
Matmerize is committed to combatting the plastics crisis through the accelerated design of recyclable, bio-based, and biodegradable polymer formulations that are simultaneously functional and economical, using machine learning and multi-property optimization.