AI & Innovation

I Transform Data into Decisions

I combine academic research, industrial experience and technical skills to design AI solutions that generate measurable value in smart cities, intelligent lighting and digital healthcare.

Areas of Expertise

Cross-functional skills spanning from research to production

Applied AI

Design and implementation of AI solutions for smart cities, intelligent lighting and predictive analytics

Data Science

Advanced statistical analysis, predictive models and big data interpretation for strategic decisions

Innovation Strategy

Strategic consulting for AI integration and emerging technologies in business processes

AI Team Building

Training and coaching for teams adopting AI-first workflows and agentic tools

Completed Projects

AI solutions that generated measurable impact

Predictive Maintenance for Public Lighting
AI system to predict failures in urban lighting systems, optimizing maintenance and improving operational efficiency.
Machine LearningIoTTime Series Analysis

Predictive maintenance

Smart City Data Platform
Integrated platform for real-time analysis of urban data (traffic, energy, environment) with predictive dashboards for decision makers.
Big DataReal-time AnalyticsVisualization

Urban data analysis

AI-Powered Energy Optimization
Optimization algorithms to reduce public lighting energy consumption based on usage patterns and environmental conditions.
Reinforcement LearningOptimizationEdge Computing

Energy optimization

Clinical Decision Support System
AI-based clinical decision support system developed during research at the University of Milan.
NLPMedical AIEvidence-Based Medicine

Published in peer-reviewed journals

Creative AI

Experiments with generative models for images and creativity

2024
Flux Training on StreetMaster
Fine-tuning experiment of Flux on StreetMaster image dataset (Cree Lighting Europe). End-to-end pipeline with Hugging Face, Replicate trainer and API. Efficiency test: 183 lm/W on LED street fixture. POC cost: ~$5 for time-to-first-result benchmark.
FluxHugging FaceReplicateProduct Imaging

POC at ~$5 - Operational best practices for brands/clients

2023
DALL-E vs MidJourney Showdown
Aesthetic perception experiment: comparison between DALL-E and MidJourney on 5 different prompts (descriptive scenes, surreal art, abstract, landscape, portrait). Community voting and analysis of style, coherence and 'vibe' differences between the two models. ChatGPT-generated feedback to interpret differences.
DALL-EMidJourneyChatGPTPrompt Design

Replicable format for workshops/universities/companies

My Approach

Iterative and results-oriented methodology

1

Analysis & Assessment

Business context evaluation, AI opportunity identification and measurable KPI definition

2

Proof of Concept

Rapid development of working prototypes to validate hypotheses and demonstrate tangible value

3

Implementation

Production deployment with integration into existing systems and internal team training

4

Monitoring & Iteration

Continuous performance monitoring, model optimization and adaptation to emerging needs

Have an AI Project in Mind?

Whether you're looking to optimize existing processes or explore new opportunities, I can help you transform AI from hype to concrete value.