A decade of Data Science and AI — scientific depth meets quantitative rigor.
From international AI stages to the Italian national track team — the moments that define the journey.
Presenting an AI platform for public-policy decisions at the Italian Parliament.
On the panel in the Parliament's press room, speaking on AI for research and innovation.
4x400m relay with the Italian senior national team at a premier Diamond League event.
On stage with LLM pipelines at scale, for one of Europe's leading data communities.
Building a 110,000-concept scientific taxonomy with graph clustering, embeddings, and LLMs.
Automating paper classification at scale with retrieval, reranking, and LLM validation.
Applied AI systems and impact-driven experimentation.
Tracking rejected manuscripts with Qdrant and Specter2, presented to the scientific Python community.
Data science models for Bitcoin and decision support.
AI initiatives and publishing innovation, presented in the United States.
AI strategy and scalable data-driven processes for editorial operations.
The azzurri lineup in Rome — the same discipline I bring to AI delivery today.
Applied AI and real-world data science impact, presented to the Swiss data community.
Applied AI for quantitative finance: models, analysis workflows, and decision systems designed for practical investing and risk-aware execution.
Professional track record in descending chronological order.
Tech Lead for Data Intelligence Team: owning roadmap and delivery of ML and LLM products, including multimillion-record semantic tagging, ontology-based graph clustering, and production pipelines with Azure OpenAI, Qdrant, and Prefect.
Led data science across a multi-billion-euro retail operation: delivered 3% revenue and margin uplift, reduced churn by 20% on 6M+ customers, improved CLV by 15%, and automated workflows with 90%+ time savings.
Delivered quantitative consulting for banking clients using SPSS, SAS, and Python across finance and risk transformation projects.
Built statistical models for credit risk and scoring, including Machine Learning, Value at Risk, Expected Loss, Loss Given Default, and Exposure at Default.
Quantitative training spanning statistics, economics, finance, and modern data science.
2021 · Advanced ML, NLP, and data products
2017 · Financial Markets, Insurance, and risk analytics
2014 · Statistical modeling and econometrics
Six areas that define my curiosity, energy, and long-term growth.
Continuous experimentation on models, LLM systems, and practical AI applications with real-world impact.
Research on market behavior, portfolio logic, and data-driven investing across traditional and digital assets.
Sharing practical lessons through conferences, mentoring, and collaboration with professionals and students.
Former Italian National Team athlete focused on consistency, resilience, and performance under pressure.
Designing efficient workflows and scalable processes to amplify execution quality and team output.
Personal interests that keep me creative and balanced, with a special place for italodance genre.
I help companies turn AI ambitions into working systems — the same LLM, GenAI and data science expertise I apply in production every day, available for your team.
Which use cases are worth building, which to buy, and which to skip. Roadmaps, feasibility checks and honest answers before you commit budget.
Design and hands-on delivery of LLM pipelines, RAG and agentic systems — with the evaluation and monitoring that make them production-grade.
From executive sessions to hands-on courses for engineering teams — the same material I bring to international stages, tailored to your context.
Forecasting, statistical modelling and quantitative finance systems, end to end: from problem framing to models your team can maintain.
A free 30-minute conversation to understand the problem — and to tell you honestly whether AI is the right answer.
A clear plan: deliverables, timeline and pricing. No open-ended engagements, no surprises.
Focused sprints or ongoing advisory, with knowledge transfer built in so your team stays autonomous.
Confidentiality by default — engagements run under NDA and client names are never published. I can walk you through approach and results in person, and references are available on request.
A curated selection of projects across machine learning, NLP, and AI product engineering.
Interactive WebGL concept map for exploring MDPI's 115,000-concept research taxonomy, built with Vue 3, deck.gl and D3 on top of SPECTER2 embeddings and ontology-based clustering. The underlying taxonomy is now public as an open dataset on Hugging Face.
Data Science framework for analyzing Bitcoin market dynamics, price patterns, and trading signals with predictive models.
View Project
Conversational AI chatbot powered by LLM technology that acts as a digital twin, answering questions about my work, projects, and career using RAG architecture and streaming responses.
Reconstructed ball trajectories across pool and carom billiards matches using computer vision: color-based detection, Kalman filter multi-object tracking with Hungarian assignment, homography for top-down court mapping, and cushion-bounce physics.
NLP pipeline aggregating sentiment from social media and news to detect Bitcoin trend reversals and support trading decisions.
View Project
Real-time scraping and data pipeline for betting exchange odds and live football statistics to power sports analytics strategies.
View Project
End-to-end pipeline for handcrafted music embeddings, genre classification with SHAP, and similarity search on an Italodance/Trance dataset.
View ProjectEnd-to-end capabilities from statistical research to production-grade AI systems.
Conference sessions, invited talks, and academic events grouped by year.
My intervention at the Italian Parliament, presenting an AI platform for public-policy decisions in research and innovation (in Italian).
How to turn millions of noisy author keywords into a five-level taxonomy of 110,000+ concepts, with SPECTER2 embeddings, Leiden graph clustering, and LLMs at targeted decision points.
Tagging millions of scientific papers against a 110,000-concept taxonomy with vector retrieval, cross-encoder reranking, and selective LLM validation.
Delivered a talk at PyData London on practical AI and data product delivery.
Presented an AI platform supporting public-policy decisions in research and innovation, at a press conference of the Italian Chamber of Deputies.
Invited session focused on applied AI and real-world data science impact.
Community session on Python, data pipelines, and practical engineering workflows.
Follow-up talk on AI experimentation and production thinking for Python practitioners.
Invited talk on data analyst and data scientist career paths.
Talk on data and analytics applications for business and engineering students.
Industry projects with graduating students focused on analytics and problem solving.
Interview about my career path as Data Scientist at Esselunga.
Books, essays, and research papers — free to read online or download.
Extending the OpenAlex taxonomy with 114,000 fine-grained scientific Concepts.
Co-authored with Feichi Lu, Oliver Grun, Mariia Eremina and Andrea Perlato at MDPI Data Intelligence. SCALE extends the OpenAlex taxonomy with a new fifth level of 113,892 fine-grained scientific Concepts, built by combining SPECTER2 embeddings, LLM-based granularity classification, and Leiden graph clustering over nearly three million MDPI author keywords. The resulting taxonomy powers Atlantis and the Taxonomy Explorer, and is already deployed in production to tag MDPI's corpus of close to two million papers.
Cos'è davvero, cosa può fare, come governarla.
A clear, jargon-free essay on artificial intelligence written for the people who have to decide about it — executives, managers and policymakers. What AI really is, what it can and cannot do, and how to govern it responsibly.
Athletic milestones from my years with the Italian national 4x400m relay teams.
Selected for the 4x400m relay at the Golden Gala in Rome, competing with the Italian senior national team in one of the premier European track events.
Member of the Italian 4x400m relay team at the European U23 Championships in Tampere, where we finished second in Europe.
For consulting, speaking, or just to talk data and AI — my inbox is open.
