Hi! Check my work
MSc AI @ CentraleSupélec | TensorFlow Certified | Computer Vision | Deep Learning
Recent Work
Heat Equation ML
12/2024
Neural network approaches for solving PDEs: CNN, MLP, and Physics-Informed Neural Networks (PINNs).
- •6 structured experiments comparing architectures
- •Physics-informed loss for better generalization
- •96x fewer parameters with function-based approach
Sub 3H Marathon Prediction
10/2023
LSTM & Transfer Learning for heart rate prediction from marathon activity data.
Defect Detection with Heatmaps
04/2023
Object detection system to identify casting defects with visual heatmap overlays.
Microstructure Classification
01/2023
YOLO-based system for automated Martensite/Pearlite identification in steel samples with FastAPI deployment.
Skills & Tools
Image Classification
Object Detection
Deep Learning & MLOps
GitHub Projects
Open source projects and academic work from my MSc AI journey at CentraleSupélec
Physics-Informed Neural Networks
Conditional PINNs solving 2D heat diffusion with systematic comparison of CNN vs MLP vs PINN architectures.
Steel Microstructure Classification
YOLO-based system for automated Martensite/Pearlite identification in steel samples with FastAPI deployment.
Heart Rate Prediction
LSTM & Transfer Learning for heart rate prediction from activity data. Deep Learning course project.
Vampires vs Werewolves AI
Strategic game AI using Alpha-Beta pruning with smart move filtering. Production ready!
Pfizer Territory Optimization
Multi-objective optimization for sales territory assignment using NSGA-II and Simulated Annealing.
Formula 1 Data Analysis
Formula 1 data analysis and visualization exploring race strategies and performance metrics.
About Me
Background
From the foundry floor to neural networks. Spent 3 years at Howmet Aerospace developing gas turbine components, where I saw firsthand how quality issues impact production. This experience drives my passion for AI-powered manufacturing solutions.
Education
Currently pursuing MSc in AI at CentraleSupélec, one of France's top engineering schools. Specialized in materials characterization through deep learning: YOLO-based systems, CNNs for defect detection, and Physics-Informed Neural Networks achieving 3.21× performance improvement.
What's Next
Upcoming internship at SAFRAN TECH POLE focusing on microstructure prediction. Looking for AI/ML opportunities in aerospace, manufacturing, and materials science. Franco-Italian, fluent in 🇮🇹🇫🇷🇬🇧 (TOEIC 910) + 🇪🇸🇯🇵.