I am a recent master graduate in Telecommunication Engineering from Politecnico di Milano with a deep passion for applying data-driven methods across diverse application domains, including communication systems, Edge-AI sensors, and biomedical applications. I am particularly interested in Explainable AI (XAI), aiming to create AI models that are transparent and interpretable for end-users and stakeholders. My expertise spans AI/ML, deep learning, and XAI, with hands-on experience in developing scalable solutions tailored to specific domains.
I recently completed an AI Researcher role at STMicroelectronics, where I developed cutting-edge solutions for MEMS calibration using TinyML and Quantization-Aware Training. My work applies XAI techniques to ensure that even complex models are understandable, whether in communication networks, medical diagnostics, or real-time sensor applications.
- 🌱 I’m currently focused on deepening my knowledge in ML/DL applications, XAI, and multidisciplinary AI/ML solutions.
- 👯 I’m eager to collaborate on AI projects that push the boundaries in communication systems, IoT sensors, and biomedical data.
- 💬 Ask me about Explainable AI, Data-Driven Systems, Deep Learning, Signal Processing, or Computer Vision.
- 📫 How to reach me: LinkedIn, Email
- ⚡ Fun fact: I’m passionate about making AI more interpretable and useful, bridging the gap between black-box models and human understanding.
Tools & Platforms: Linux, Git, Docker, Kubernetes (familiar), AWS (Amazon EC2), Jupyter, Anaconda, Pyenv, VS Code, SQL
Libraries: TensorFlow, PyTorch, scikit-learn, XGBoost, LightGBM, SHAP, LIME, GradCAM, OpenCV, PySpark, Matplotlib, Pandas, NumPy
Developed a system to accurately detect and count vehicles in high-density environments utilizing YOLOv8 and BYTETrack. This project demonstrates the use of computer vision in real-time environments, applicable in smart transportation systems.
Implemented machine learning models to detect anomalies in optical transponder signals using constellation diagrams. This project applied ML techniques to improve communication systems, ensuring data integrity and network reliability.
Conducted an extensive analysis comparing classifiers using radiomics and features extracted from deep neural networks for biomedical image analysis, with a strong focus on Explainable AI (XAI) techniques (such as GradCAM, SHAP, and LIME) to ensure transparency in the diagnostic process.
Performed an in-depth analysis of performance indicators across 20 companies, leveraging advanced statistical techniques to understand market trends and optimize e-marketing strategies. This project was proposed by Google Italy.
Here are some of the main courses I completed during my MSc in Telecommunication Engineering at Politecnico di Milano:
- Network Measurement and Data Analysis Lab (30/30 Cum Laude)
- Fundamentals of Signals and Transmission (30/30 Cum Laude)
- Applied AI in Biomedicine (30/30)
- Recommender Systems (30/30)
- Game Theory (29/30)
- Wireless Internet (28/30)
- Image Analysis and Computer Vision (27/30)
- Advanced Digital Signal Processing (27/30)
-
IMU Self-Calibration by Means of Quantization-Aware and Memory-Parsimonious Neural Networks
Published in MDPI Electronics Journal, October 2024. -
IMU User Transparent Tiny Neural Self-Calibration
Published at the 8th International Conference of Research and Technologies for Society and Industry (IEEE RTSI 2024). -
Continuous MEMS Self-Calibration Process by Means of Tiny Neural Networks
Accepted at STMicroelectronics TechWeek 2024 internal conference and submitted as an innovative solution to enhance MEMS calibration system architecture, March 2024. -
An Introduction to Convolutional Neural Networks & Applications
Published as a poster presentation at the 4th National Conference on Contemporary Issues in Computer Information and Science (CICIS2019), January 2019.
MSc in Telecommunication Engineering
Polytechnic University of Milan | Sep 2021 - Jul 2024
Specialization: Signals and Data Analysis
Thesis: Continuous IMU-MEMS Self-Calibration Process by Means of Tiny Neural Networks
BSc in Computer Science
Kharazmi University of Tehran | Sep 2014 - Jul 2020
Thesis: Automatic Architecture Design of CNNs using Genetic Algorithm and Reinforcement Learning (MetaQNN)
- Member of the PoliMi Data Science Association, contributing to various ML and data science projects.