cv

Head

Institute for Decentralized AI

• Head of a research institute focused on security and infrastructure for decentralized and distributed AI systems.
• Two main units: Oxford and Stanford.
• Funded by the AI Safety Fund and the Cosmos Institute.

PhD Research Intern

Microsoft

• PhD internship in Microsoft's Agents Team of the Applied Sciences Group.
• Main focus: developing efficient local agents.

Predoctoral Research Fellow

University of Bologna

• Department fellowship on adversarial robustness under Prof. Michele Lombardi.
• Proved that adversarial defenses for neural networks are Σ^P₂-hard (i.e., computationally infeasible).

• Samuele Marro, Jialin Yu, Emanuele La Malfa, et al., Benchmarking at the Edge of Comprehension, ICML 2026 (Oral).

• Emanuele La Malfa, Gabriele La Malfa, Samuele Marro, et al., Large Language Models Miss the Multi-Agent Mark, NeurIPS 2025 (Oral).

• Tobin South, Samuele Marro, Thomas Hardjono, et al., Authenticated Delegation and Authorized AI Agents, ICML 2025 (Oral).

• Samuele Marro, Davide Evangelista, Emanuele La Malfa, et al., Language Models Are Implicitly Continuous, ICLR 2025.

• Samuele Marro, Michele Lombardi, Computational Asymmetries in Robust Classification, ICML 2023.

Grant, 550k GBP

UK Advanced Research and Innovation Agency

• Grant by the Scaling Trust division of the Advanced Research and Innovation Agency on automated algorithmic game design.

Grant, 450k USD

AI Safety Fund

• Grant by the AI Safety Fund (joint Google-OpenAI-Anthropic-Microsoft fund) on the decentralized oversight of agent networks.

Microsoft Studentship, 250k USD

• Studentship for research into communication protocols between LLMs and multimodal memory in generative models.

DPhil in Artificial Intelligence

University of Oxford

• 4-year DPhil at the Oxford AIMS CDT under Prof. Philip Torr on multiagent LLMs.
• Funded by a 250k USD Microsoft studentship. Graduation: Oct 2028 (can be anticipated).

MSc in Artificial Intelligence

University of Bologna

• GPA: 29.75/30, top of the class (170 students). Graduation grade: 110 cum laude.

Excellency Degree (COTS)

Institute of Superior Studies

• 5-Year excellency degree (EQF8, Italian system). Equivalent to a Master's Degree.
• GPA: 29.88/30. Research topic: adversarial robustness.

BSc in Computer Science

University of Bologna

• GPA: 29.77/30, top of the class (150 students). Graduation grade: 110 cum laude.

Exploration Grant, 20k GBP

UK Advanced Research and Invention Agency

• Exploratory grant on automated protocol design as a path to full cybersecurity resilience.
• Part of the "Trust Everything Everywhere" research area.

AIxIA "Leonardo Lesmo" Award

• Award for the best MSc thesis in AI in Italy, granted by the Italian Association for Artificial Intelligence.
• I accidentally submitted my BSc thesis, but won anyway.

Hackathon Prizes, 7k USD

Gitcoin Hackathons

• 6k USD won by developing Soliditypp smart contracts for the Vite blockchain.
• 1k USD won by developing tooling to improve IPFS usage.

CINECA HPC Grant, 170k Core-Hours

• Grant for the Galileo100 HPC cluster for research on the computational complexity of adversarial attacks.

• Samuele Marro, Philip Torr, LLM Agents Are the Antidote to Walled Gardens, ICML 2026.

• Samuele Marro, Alan Chan, Xinxing Ren, et al., Permission Manifests for Web Agents.

• X Angelo Huang, Emanuele La Malfa, Samuele Marro, et al., A notion of complexity for theory of mind via discrete world models, EMNLP 2024.

• Samuele Marro, Emanuele La Malfa, Jesse Wright, et al., A Scalable Communication Protocol for Networks of Large Language Models.

• Andrea Asperti, Davide Evangelista, Samuele Marro, et al., Image embedding for denoising generative models, Artificial Intelligence Review.

Fellow

Cosmos Institute

• Research fellow on the intersection of AI, philosophy, and decentralization.
• Goal: build stable non-governmental institutions to steer AI towards a less authoritarian future.

Solidity Developer and Auditor

Freelance

• Performed Smart Contract audits, Web3 integrations and secure Smart Contract development.
• Specialized in EVM and EVM-like (e.g. Vite) security standards.

Machine Learning Intern

MindIt Srl + Kantar

• Developed a Tensorflow LSTM implementation compatible with IBM CPLEX optimization algorithms.
• Rewrote Meta's FBProphet time series model in Tensorflow.

Teaching Assistant

University of Bologna

• Taught the Applied ML module of "Introduction to Machine Learning" for the Fall 2022 semester.

Software Development Intern

Sistemi Tre, Alba, Italy

• Automated database queries through the in-house software interface using AHK, leading to a 3-4x increase in data retrieval speed.
• Assisted IT staff during on-site technical support.

Chair

Lightweight Agent Standards Working Group

• Industry-academia joint working group, focused on designing and iterating quickly on standards for agent interactions.
• Notable organizations: Oxford University, Stanford University, MIT, Firecrawl.

Mentor

Cambridge ERA Fellowship

• Mentored fellows on applications of multi-agent security techniques to AI safety.

Head

Computer Science Student Commission, Superior College of Bologna (ISS)

• Supervised CS student projects and advised students on CS and ML-related research directions.

Representative

CS Department, University of Bologna

• Managed student-department relations and represented student interests in department meetings.