About Navexis
We started Navexis after watching talented chemists spend days synthesising and testing compounds that a good prediction could have ruled out in seconds.
Our mission
Drug discovery runs on a simple, expensive loop: design a molecule, make it, test it, learn, repeat. Every turn costs days of lab time and thousands of dollars, and most molecules don't make it.
Navexis shortens the loop. Our machine learning models predict the physical properties that govern a compound's behaviour, beginning with solubility and lipophilicity, before synthesis. A proprietary, data-efficient approach reaches reliable predictions from fewer measurements, and the models sharpen to each lab's own data, conditions, and chemistry.
We believe this capability shouldn't only exist inside the largest labs. A small biotech deserves the same predictive power as a lab ten times its size, and our long-term vision is an intelligent decision-making partner for every chemist, with the scientist always in the loop.
Where we are today
We're an early-stage team based in Melbourne. Our first models cover solubility and lipophilicity, two practical, high-value properties in early drug discovery, with the framework built to expand across the wider property picture.
Beyond drug discovery, the same approach extends to other chemistry-led fields where property prediction matters, including formulations, cosmetics, materials, and environmental science.
We're currently speaking with investors, research partners, and early users. If you'd like to be one of them, get in touch.
// The team
Founders
Theodore Seng
CEO & Co-founder
Theodore leads Navexis: strategy, partnerships, fundraising, and operations. He drives the company's mission to bring rigorous, data-efficient property prediction to laboratories well beyond the largest companies, and is the first point of contact for investors and partners.
Viktor Prypoten
Scientific co-founder
Viktor leads the science at Navexis. A computational chemist specialising in machine learning for molecular property and binding prediction, his peer-reviewed research in the Journal of Chemical Information and Modeling spans molecular dynamics methods for predicting binding free energy and machine-learning design rules for membrane permeability. He builds the predictive models at the core of the platform.
Navigate every molecule with confidence.
Navexis