Complex functional materials and devices sit in a regime mainstream ML-for-science rarely touches: the systems are complicated, every data point is expensive, and there is no database or benchmark to start from. That regime is what my research targets, with an AI co-scientist stack running from hypothesis generation through device digital twins to physical validation.
The organizing idea: autonomy in this regime is a decision problem first. The bottleneck is knowing which experiment is worth running, and whether the model that proposed it can be trusted. My bias, learned from running the loop by hand, is to automate the step that returns the most information per run rather than the step that returns the most runs per day.
Public databases and benchmarks cover a thin slice of materials science: mostly single crystals, small molecules, idealized surfaces. The systems I work on (FET sensors, membranes, electrocatalytic assemblies) are multi-component and strongly coupled across scales, and one data point can cost days of synthesis and testing. The big data is not coming. So I design AI systems that work the way experimentalists do: pull priors from the literature, build physics-constrained surrogates, and spend the experimental budget where it counts.
Text: an agentic pipeline, prompt-optimized with TextGrad, extracts structured knowledge graphs from raw publication corpora (21.8% BLEU improvement in knowledge extraction). Twin: a graph neural network that knows the device topology, trained as a digital twin of the coupled material–device system; on FET sensors it predicts sensitivity with 92.3% accuracy. Translation: the twin then screens 123.2 million PubChem compounds for out-of-distribution tasks such as PFAS-sensing probe design. Published in the KDD 2026 proceedings (AI4Science Track, doi:10.1145/3770855.3819013); Spotlight Oral at ICLR 2026 AI4Mat.
Text
AI reads the literature and extracts structured device knowledge.
Twin
The full device context becomes a topology-aware graph.
Translation
The twin ranks new candidate materials and molecules.
The Twin is a message-passing GNN over the device topology graph: $$h_v^{(k+1)}=\phi\Big(h_v^{(k)},\ \bigoplus_{u\in\mathcal{N}(v)}\psi\big(h_v^{(k)},h_u^{(k)},e_{uv}\big)\Big)$$ where edges $e_{uv}$ encode physical couplings between material, channel, and electrode nodes.
Deep Tree of Research (DToR) is a hypothesis-generation engine, not a chat agent: a local-first RAG system with a tree-structured orchestrator that adaptively expands and prunes research branches for coverage, depth, and coherence. Benchmarked on 27 nanomaterials/device topics against 44 agent configurations, its reports achieved a ~79% mean pairwise win rate against commercial deep-research systems, while running entirely on in-house, consumer-level hardware with open-source LLMs.
RAPIDS is the atomistic validation engine of the stack. It benchmarks machine-learning interatomic potentials against DFT on 5,567 probe–target dimer interactions across 18 benchmarks; the result is that geometry, not the energy surface, drives the neutral MLIP–DFT gap. It is also packaged as a tool autonomous LLM agents can call for fast physical sanity checks.
The last gate is the one I run myself: nanomaterial synthesis, device fabrication, and electrochemical testing at the bench. I have taken a full campaign through this loop by hand: literature mining to active learning to synthesis to a membrane-electrode assembly that outran a commercial catalyst (Science Advances, 2025).
Not every candidate deserves full-cost validation. Drag the fidelity up and watch the field thin out.
A discovery loop is only as durable as the plumbing under it. I am the lead designer and developer of the data platform for NSF MADE-PUBLIC (FMRG #2037026): a browser-accessible layer over Globus-based data lifecycle management, searchable metadata, interactive knowledge graphs, Bayesian experiment design, and simulation workflows. Collaborators across institutions can use it without owning the stack. The same instinct shapes the rest of the work: RAPIDS ships as a verification service agents can call, and DToR and T3 ship as open repositories rather than screenshots in a paper.
Electrocatalysis
Accelerated discovery of OER/HER catalysts for clean hydrogen production.
Chemical Sensing
Neuromorphic approaches for FET sensor design and environmental monitoring.
Water & Environment
PFAS detection probes and AI-guided design of PFAS-selective membrane interfaces.