Research
Preprint and commissioned work. All five preprints below are single-author and available in full on arXiv. Citation counts are tracked on my Google Scholar profile.
Argues that observational and interventional data given to a strong enough predictor is not sufficient for counterfactual reasoning. Reframes a world model as a positive semidefinite coupling kernel over admissible possible worlds, where the diagonal is ordinary prediction and the off-diagonal carries the cross-world dependence a standard predictor cannot represent.
A verifier that decides whether a tool-using agent should execute a proposed action, by mapping it to a structural causal query over a committed action-state graph and returning an auditable verdict: execute, reject, experiment, or abstain. Reports zero observed false executions under moderate and adversarial confounding.
An open-source Rust system pairing LLM-driven ontology construction with formal OWL reasoning and alignment over the Model Context Protocol. Stable one-to-one matching is shown to be the dominant factor in alignment quality, reaching F1 of 0.832 on the OAEI Anatomy track.
Treats a deterministic event graph as the substrate of a world model, so that counterfactual questions become graph queries with defined answers rather than sampled generations.
592 runs across three toy substrates plus two real-world replications, testing a mean-field model of saturating power-law growth. Finds that growth dynamics are substrate-conditional and do not generalise across domains.
Published by the Welsh Government on 17 March 2026 as part of its land value tax evidence programme. Tests five valuation approaches, including market-based statistical valuation and machine learning applications.
Research on reinforcement-learning language agents in emulated networks for threat detection, carried out under a formal research services agreement with the Alan Turing Institute. Report under review.
Open source
Tools I build and maintain in the open.
AI-native ontology engine. Rust MCP server with OWL 2 DL reasoning, SHACL validation, and SPARQL. Single binary, no JVM.
Trust infrastructure for AI agents. Provenance, integrity, and zero dependencies, in pure C.
Score any document and prove every claim, with ontology-grounded verification.
Benchmarks for world-model structure, including an off-diagonal kernel witness.
A certified axiom bridge between the Information Exchange Standard and HQDM for defence data.
Tooling for building and transforming Model Context Protocol servers.
Experience
Waste and Resources Action Programme (WRAP)
Fintech Scotland, 2026 cohort
Kalgera. Ontology-based financial crime and transaction risk scoring.
Innovate UK BridgeAI
Welsh Government
National Digital Twin Programme
Etiq AI. Adversarial red-teaming of retrieval-augmented generation pipelines.
The Alan Turing Institute. Large language models for cyber defence.
University of the Arts London
The Tesseract Academy
For research collaborations, speaking, advisory work, or anything involving ontologies and verifiable AI.