GenAI Product Engineer @ GenAI Lab - Nexi Group
With an M.Sc. in Engineering from Politecnico di Milano, I’ve spent the past year leading cross-functional teams to transform complex challenges into real enterprise solutions, from idea to rollout.
Originally from Milan (Italy), I'm a Biomedical Engineer turned GenAI innovator, driven by the belief that purposeful technology can make life more human.
After earning my M.Sc. in Biomedical Engineering (Biomechanics and Biomaterials) at Politecnico di Milano (2023), I began my career in a cutting-edge research lab, working with innovative biomaterials and complex prosthetic systems for cardiac valve repair.
Today, I'm part of GenAI Lab, the team driving GenAI implementation in Nexi Group (10k+ employees across 25 countries). My goal is simple: to help people work smarter, faster, and with less stress through the power of GenAI.
Beyond developing and delivering real-world solutions, I mentor teams on how to integrate AI into daily workflows. In 2024, I received Nexi’s internal #MakeTheDifference award for driving measurable innovation and impact.
Since 2023, I co-lead Calcio Seduto (or Seated Football), a non-profit initiative that brings children with disabilities closer to sports through an inclusive sport we invented.
What I’ve been working on
€250K enterprise HR chatbot powered by a hybrid RAG architecture, achieving 80% answer accuracy and reducing overall HR ticket volume by 25% (out of 10k+ tickets/year).
This project presents an enterprise HR chatbot designed to enhance employee support experience across a workforce of over 3,500 people. The system operates within a complex enterprise multi-tenant infrastructure, seamlessly integrating into the company’s digital ecosystem through a custom Microsoft Teams application. The backend is developed in Python and the frontend is built in React. It is compliant with the European Accessibility Act (EAA), meeting WCAG 2.2 Level A standards, and operates under legally approved Terms and Conditions to ensure regulatory alignment.
Backend and frontend are containerized and deployed on a scalable cloud infrastructure. The chatbot is accessible to users through Microsoft Teams in a multi-tenant enterprise architecture.
Successfully delivered within an enterprise multi-tenant infrastructure, achieving 80% response accuracy and a 25% reduction in HR tickets, with an annual ROI equivalent to 1.5 FTE.
GenAI meeting minutes bot built with Microsoft Copilot Studio, estimating an annual saving of over €1M and 21k hours of work.
This project introduces a meeting minutes chatbot that automates the capture and organization of meeting notes. To accelerate time-to-value and avoid complex enterprise integrations, I developed a low-code solution using Microsoft Copilot Studio, enabling a pilot-ready MVP within one month. The chatbot leverages Microsoft Teams' built-in transcript recording to generate structured, customizable summaries. It operates through a single LLM call with a dedicated system prompt, delivering a lightweight yet scalable alternative to premium tools such as M365 Copilot or Fireflies.ai. The projected annual impact exceeds €1M in savings and 21,000 hours of productivity gain. Currently in pilot phase with 35 users, following C-level approval after the successful presentation of the business case and MVP.
As Product Manager, I owned the end-to-end delivery of the chatbot, from concept to pilot launch. I led a team of 2 engineers and was responsible for the product strategy, roadmap, and technical implementation. Hands-on, I configured environments in the Microsoft Power Platform, designed the data and conversation architecture, set up the billing policy, and managed stakeholders. I pitched the business case and MVP to C-level executives, securing approval for the pilot rollout.
Developed entirely in Microsoft Copilot Studio, the solution was deployed in a controlled staging environment for pilot validation.
Experimental optimization of the industrial manufacturing process for cardiac valve prostheses, focusing on the dehydration cycle of pretreated bovine pericardium.
This project presents the optimization of the manufacturing process of fixed bovine pericardium used in biological heart valves. Design of Experiments (DoE), thermo-mechanical testing, and accelerated aging were employed to define a repeatable, scalable, and regulatory-compliant process for Class III medical devices. The key outcome was the identification of a glycerol-based treatment that preserves the tissue's physical properties while enabling dry storage, simplifying manufacturing compared with the state-of-the-art "wet" process.
The identified process improved operational efficiency (no liquid processing, no aldehyde sterilization, simplified storage and logistics, and reduced process time) while maintaining the tissue's functional properties, surpassing state-of-the-art results.
Technical expertise and professional skills
Yes, I built this site myself! I learned by doing with a mix of YouTube tutorials, general programming basics, and a healthy dose of GitHub Copilot. I did use a bit of AI help, but fun fact: there are no em dashes on this page. In the footer, you’ll see it’s “optimized for desktop”, which is just a polite way of saying I’m not yet good enough to design a mobile version I’m proud of.
The Nexi Graduate Program I joined is a 12-month career accelerator featuring two 6-month rotations across key strategic departments. It’s designed for top graduates to develop cross-functional experience, expand both business and technical skills, and gain a full-spectrum view of the FinTech industry. The program selects from a competitive pool of 3,000+ applicants with <1% acceptance rate. You can read a short post about my experience in the program here: The #FutureShapers Experience!
I see Generative AI not as a novelty to romanticize, but as a toolkit that gives people superpowers: faster prototyping, clearer drafts, and new ways to solve real problems. As an engineer, I’m fascinated by how science impacts reality, which probably explains why I’ve recently been exploring quantum mechanics and its implications for computation and algorithms. In my free time, I also authored a published piece for Nexi on GenAI trends in 2024: read the article (yes, it's in Croatian... and no, I don't speak Croatian).
Huge fan of sports! I grew up playing football (soccer) and I still play regularly. I’ve also coached youth teams. I plan to run my first marathon in 2026.
No.
