The engine is open source. Clone it, or book a walkthrough for cloud deployment.
Book a walkthrough →Research
Published papers, preprints, and internal research.
FAVES V4: A Production-Deployed Compliance and Liability Screening System for Drug Discovery at 122M-Compound Scale
Aligned to HHS/ONC FAVES and NIST AI 600-1. 362/362 compounds passing post-fix; ~5 ms per molecule live, 122,407,635 documents cached.
Ari Harrison
Protocol-Mediated Execution Governance for LLM Agent Systems
Security and governance for MCP-based scientific agents: protocol-level execution control, token exchange, and an immutable audit trail.
A. Harrison
NovoExpert: State-of-the-art ADMET prediction on TDC benchmarks
Five SOTA wins on Therapeutics Data Commons ADMET endpoints using CatBoost ensembles with MapLight and GIN fingerprints.
Ari Harrison
From 2D Graphs to Sub-Unit pKa: Nine Iterations to Beat MolGpKa on Both Scaffolds and Sulfonamides
A routed-ensemble pKa predictor with functional-group-specific atom-level descriptors — a v1→v9 ablation and competitive analysis.
A. Harrison
FAVES V3: A Computational Framework for Evaluating Molecular Compliance Across Regulatory Jurisdictions
Three-tier detection system for controlled substance screening across 8 jurisdictions. 100% sensitivity and specificity on a 102-compound benchmark.
Ari Harrison