Company

AI for the chemistry of liquid mixtures.

Liquid mixtures power batteries, drug delivery, and specialty chemistry — yet most ML stops at single molecules. Solvolyte is built around a different premise: that the mixture is the unit of design, that geometry governs transport, and that a model is only as good as the bench experiment it predicts. We engage as paid 8-week design partnerships with R&D teams who need a shortlist, not a paper. Every engagement closes the loop between prediction and electrochemistry, so the model your team uses in week eight is sharper than the one it started with.

01

Physics-first

SE(3)-equivariant message passing enforces rotational and permutation symmetry by construction. Solvation-shell geometry, ion-pairing distances, and conformer ensembles are inputs — not learned from scratch. The model never violates the underlying physics, even out-of-distribution.

02

Experiment-grounded

Every shortlist is closed-loop with a synthesis and electrochemistry partner. Predictions ship with calibrated uncertainty, and the experimental return signal updates the next training round. We don't ship a number we wouldn't bet a Coulombmeter on.

03

Open by default

Pre-trained GeoSolv weights, the Set-SE(3) reference implementation, and the CheMixHub-derived training pipeline are released under permissive licences upon paper acceptance. Partners keep full IP on derived predictions; we keep the method open so the field moves forward.

Founders

Physics meets learning.

QH

Qi-Nan (Eric) Huang

Co-Founder · CEO

Agentic AI for Electrolyte Discovery

PhD student at the Pritzker School of Molecular Engineering, UChicago (Amanchukwu Lab). NSF AIMEMS NRT trainee. Lead author on the GeoSolv framework. B.S. Chemistry, Hunan University.

Click to view LinkedIn →

IS

Ian Su

Co-Founder & CTO

ML Systems & Training Infrastructure

Builds the distributed training stack and screening pipeline that turns 100k CheMixHub datapoints into a 10⁵-candidate ranker. Background in large-scale ML systems and scientific computing.

WW

Weizhou Wang

Co-Founder & Head of Experimental Validation

Electrochemistry & Experimental Validation

Closes the loop between the model and the bench — designs synthesis and electrochemical characterization protocols for the high-conductivity Na-ion shortlist.

WX

Weizhi Xue

Co-Founder & Head of Molecular Dynamics

Molecular Dynamics & Solvation Physics

Drives the physics-constrained side of GeoSolv — molecular-dynamics studies of solvation-shell geometry and ion-pairing that ground the equivariant network in real liquid behavior.

Click to view LinkedIn →

AC

Adrian Cao

Co-Founder & Head of Architecture

Equivariant Architectures

Designs the Set-SE(3) message-passing core. Specializes in 3D-equivariant neural networks for atomistic systems and group-theoretic inductive biases.

Click to view LinkedIn →

Contact

Let's scope a pilot.

Partnerships

[email protected]

Based in

Chicago, IL · Pritzker School of Molecular Engineering

Open science

Code and pre-trained GeoSolv models will be released open-source upon publication.