PhD in Artificial Intelligence · Politecnico di Torino · Milan, Italy

Mario Trerotola

Machine-learning systems whose outputs can be audited: which source, which feature, which rule. The decision stays with a person.

Financial supervision is where I work, because it is one of the hardest places to make these systems hold: adversarial data, rules that are written down, and someone who has to answer for the outcome. From 2024 to 2025 I built one inside Italy’s financial markets authority, on an institutional placement from my doctorate; before that, co-founder and CTO of a PropTech backed by Exor Ventures; now a researcher at the University of Salerno. The methods are not specific to finance: model-agnostic explanation, an agent runtime, AI Act patterns for high-risk systems.

Open to research collaboration, applied roles, and building with people who have to stand behind what their systems output.

Research

08 · 2026

Sperimentazione dell'IA nell'azione di contrasto della CONSOB agli abusivismi finanziari

Using Artificial Intelligence to enhance CONSOB’s enforcement against unauthorised financial activities

M. Trerotola, A. Boghi, P. Deriu, G. Frega · Consob, Quaderni FinTech, No. 19

A hybrid prototype — language models with deterministic rules — for investor complaints and the websites of reported entities, developed by Consob with the Politecnico di Torino and the Università Politecnica delle Marche. Written with the heads of the authority’s Studies and Regulation Division and of its Unauthorised Phenomena Supervision Office.

The system takes no decision on its own: it prepares, and the assessment, the decision and the responsibility for it stay with the operator.

consob.itJEL O32, O33, C55, G21

Financial crime and market integrity

07 · 2026

Blockchain Fraud Detection for Explainable Forensic Investigations

M. Trerotola, D. Calvaresi, M. Parente · DSA ISC 2026

Deep models are accurate but opaque; interpretable pipelines are usually tied to one chain. This method keeps both: architecture-neutral graph-temporal features, Balanced Random Forests for severe class imbalance, and a three-layer XAI stack (SHAP, CIU, DEXiRE). Macro-F1 up to 0.92 on 31,000 Bitcoin wallets and 0.985 on 157,000 Ethereum tokens, with the reasoning left inspectable.

Explainability and audit

04 · 2026

A Hybrid Multi-Agent System for Early Scam Detection in Crypto-Assets

M. Trerotola, M. Parente, D. Calvaresi · Applied Sciences, 16(7), 3122

Seven specialised agents — coordination, retrieval, heuristic risk, MiCAR compliance, on-chain fraud detection, reconciliation — combining language models with rule-based compliance checks. Component-level testing on 150 projects gives 95% output reproducibility across five reruns and 210 s mean latency: proof of concept, not deployment evidence. A pilot with six researchers and two regulatory-authority experts supplies preliminary usability findings.

doi:10.3390/app16073122 code

Agent systems

The full record: 9 papers on ORCID, 14 repositories on GitHub.

Project

InvestShield watches the public stream of TLS certificates and captures a dated, hash-sealed copy of a site imitating a bank or an investment firm — while it is still online, before the complaint that would otherwise be the first alert. My own project, live.

How it works

Ventures

Alongside the research, as a builder: co-founder and CTO of a venture-backed company, and a working prototype built at MITdesignX.

2025

Venture Fellow, MITdesignX Venice

Selected for the Venice cohort of the venture programme of the MIT School of Architecture + Planning. Built the first working prototype of an AI-driven interior fit-out platform, using generative models for automated space design.

2023–2025

Co-founder and CTO, Rellai

PropTech backed by Exor Ventures through the Vento programme and incubated at I3P, the incubator of the Politecnico di Torino. Built the product, the cloud infrastructure and a payment platform tying renovation payments to verified project milestones. Web Summit Startup Programme; Klimahouse 2025 Startup Contest. The company closed in 2025.

2023

Entrepreneur in Residence, Vento

Selected among 40 participants for the venture-building programme funded by Exor Ventures. Formed the founding team, pitched the Exor investment committee and secured pre-seed funding.

Contact

For research collaboration, an applied role, or building something together. Finance is where most of my cases have been; the problem — a system that has to justify its decisions to the people accountable for them — is not specific to it.

mariotrerotola@gmail.com