I’m Saverio Mattia Merenda, an AI Solution Architect who holds a Master’s Degree in Computer Science (110/110 cum laude) from the University of Parma, Italy.
My academic and professional path sits at the intersection of artificial intelligence, software engineering, and software security. Building on a Master’s focused on Artificial Intelligence and Software Verification, I actively design and engineer advanced AI systems, combining research-driven methodologies with production-grade architectures.
Currently, I work as an AI Solution Architect, where I design, develop, and deploy custom AI agents and AI-driven systems for both client-specific solutions and internal optimization projects. My work includes the development of autonomous agent ecosystems, continuous-learning AI agents for customer support, and meta-agent architectures capable of automatically configuring and generating other AI agents. I also engineer ad-hoc MCP tools and internal AI tooling platforms to support agent orchestration, lifecycle management, and scalable deployment, integrating these systems into enterprise workflows through modern automation and orchestration platforms.
In parallel, I am deeply involved in academic research and advanced software development. I contributed to the development of EVMLiSA, a static analyzer for EVM bytecode based on abstract interpretation, which I extended to cross-chain ecosystems and equipped with a static checker for Access Control Incompleteness in cross-chain bridges, the subject of my Master's thesis. My research activities include software security, access control analysis in smart contracts, and the integration of large language models with static analysis to improve the quality and reliability of AI-generated code. These activities have led to peer-reviewed publications and participation in international research projects.
Alongside research and industry work, I have served as a Teaching Assistant at the University of Parma, delivering academic lectures and supporting undergraduate education in computer science. This role strengthened my ability to combine theoretical rigor with practical system design, communication, and mentorship.
I am driven by the construction of intelligent systems that are not only functional, but also scalable, verifiable, and architecturally sound. My focus is on building AI infrastructures that merge automation, learning, and formal methods, with a long-term vision toward reliable, autonomous, and secure AI-driven systems.
You can find my full CV here.
Education
See all (2) →Thesis title: Static Detection of Access Control Incompleteness in EVM Smart Contracts: Relational Taint Analysis for Cross‑Chain Bridges. · University of Parma, Italy · 110/110 cum laude
Thesis available
Thesis title: Construction of sound Control‑Flow Graphs for bytecode EVM. · University of Parma, Italy · 108/110
Thesis available
Experience
See all (6) →Keplero AI · Milan, Italy
University of Parma · Parma, Italy
Publications
All (2) →Special issue of: Novel Applications of Blockchain for Information Systems and Business Process Management (1 September 2025) · Journal paper · 30 August 2025 · Open access
PDF available
V. Arceri, S.M. Merenda, G. Dolcetti, L. Negrini, L. Olivieri, E. Zaffanella. Towards a Sound Construction of EVM Bytecode Control-flow Graphs26th ACM International Workshop on Formal Techniques for Java-like Programs (FTfJP 2024) · Workshop paper · 25 July 2024 · Vienna, Austria
PDF available
Talks
All (1) →26th ACM International Workshop on Formal Techniques for Java-like Programs (FTfJP 2024) · Conference presentation · 20 September 2024 · Vienna, Austria
Slides available · Attendance available
Research Projects
All (1) →LLMs Meet Static Analysis: improving quality and reliability of AI‑generated code · 2024 · University of Parma, Italy
Events
All (3) →Conference participant · Symposium · Venice, Italy
Student · Summer School · Lipari, Italy
Attendance available