Founded Axine to apply machine learning to two problems he cares about: soil contamination and drug discovery. Also leads the foundation's global healthcare work.
Built on precision,driven by impact.
Axine builds ML systems for two problems: mapping soil contamination from hyperspectral satellite data, and ranking drug candidates against specific disease targets. We work in environmental science and biomedical research because both fields have urgent, unsolved problems where better data analysis changes outcomes.
Our Vision
Contaminated land gets identified and prioritized before it harms people. Drug candidates that fail for the wrong disease context stop wasting lab time. Scale and speed of scientific assessment catches up to the scale and speed of the problems.
What We Do
We use machine learning and computational modeling to solve problems in environmental monitoring and drug discovery that don't respond well to traditional approaches.
AxineHMAS reads hyperspectral satellite data to predict soil heavy metal concentrations across large areas — no field sampling needed. It turns what used to be weeks of lab work into a spatial map produced remotely.
01AxineScreen generates and scores drug candidates for a specific disease target. The Drug Worthiness Score combines ADMET profiling, toxicity prediction, and selectivity into a single priority ranking — so researchers spend lab time on compounds that are actually likely to work.
02Both platforms reduce the cost of being wrong early. Whether that's sampling the wrong fields or advancing the wrong compound, better prediction earlier in the process saves significant time and money downstream.
03Why Choose Us
Our team includes computational scientists, AI researchers, chemists, and environmental specialists. The problems we're working on need all of those perspectives in the same room.
We use machine learning, generative modeling, and in silico simulation — not because they're fashionable, but because they're the right tools for problems where physical experiments are too slow or too expensive to run at scale.
Our models are validated against real datasets and built with disease-specific and environment-specific context baked in. We don't ship a generic model and call it done — the science matters to us.
From contamination mapping to drug candidate prioritization, our platforms are designed to change what researchers can do in a day — not just add a new dashboard to an existing workflow.
Meet the Axine Team
Axine's team spans computational science, AI research, chemistry, pharmacology, and environmental science. We're a small, technical group that moves fast and takes the underlying science seriously.
Our work covers machine learning, generative chemistry, hyperspectral imaging, pharmacokinetics, and ADMET modeling — across two platforms built to solve real problems in drug development and environmental monitoring.
Founded Axine to apply machine learning to two problems he cares about: soil contamination and drug discovery. Also leads the foundation's global healthcare work.
Leads the pharmacological science behind AxineScreen — making sure the drug scoring models reflect how compounds actually behave in the context of a target disease.
Shais brings a relentless drive and infectious energy that elevates everyone around him. His commitment to excellence and hunger to win make him a force at the commercial frontier of Axine.
Pharmacologist specializing in drug mechanisms and ADMET profiling. Works directly on molecular evaluation for AxineScreen.
Keeps research and engineering moving — manages workflows, timelines, and the coordination between teams that makes it all work.
Builds and maintains Axine's AI systems — including the generative modeling pipelines and core platform infrastructure for both products.