Machine Learning Engineer and passionate about Space, based in Rome, Italy

Email / LinkedIn / GitHub

Machine Learning Scientist @ PiSchool (March 2023 - Present)
Main duties: I lead international teams of engineers in solving real business problems, driving projects based on state-of-the-art research in Machine Learning. I actively contribute to hands-on development, working closely with the team. I ensure successful outcomes, through my expertise in fast prototyping, knowledge of Scrum (Agile) principles, and staying updated on AI trends. Moreover, I played a pivotal role in securing and advancing major European tenders.

Machine Learning Specialist @ Eyedea Recognition (Nov 2022 - March2023)
Main duties:

• Design,development and implementation of Machine Learning algorithms
• Data Management
• Software application development
  
  

Software Engineer YGT @ ESA (Sep 2021 - Oct 2022)
Main duties:

• Detailed analysis of the data access flow for Earth Online users
• Content synchronisation analysis for large system of systems
• Sustainable systems and earth friendly hosting, operation and maintenance in IT
• Correlation of user numbers from different systems
• Analysis on the equivalent products for data analysis and storage to be used
• Optimization of EO-CAT data visualization
• Solutions to exploit knowledge graphs to esa web-pages
• Development of AI based web-page using state of the art Deep Learning models
  
  

Deep Learning Research Intern @ INRIA (March 2021 - Aug 2021)
The project has been developed during the internship under the supervision of Emanuele Natale and Andrea Clementi. The internship aimed at tackling the problem of compressing artificial neural networks via iterative pruning approaches. I reviewed state-of-the-art approaches, devised variants of known methods and designed novel approaches, and extensively validated those methods against known ones. Iterative pruning methods have been a classical approach for neural network compression since several decades. While earlier methods relied on heuristic arguments, such as assumptions on the Taylor approximation of the loss function, recent approaches have attempted to provide rigorous guarantees by leveraging algorithmic techniques. The project focused on the assessment of the merits and shortcomings of such recent contributions. The proposed implementation language for the project is the Julia language.

• Technologies used: Julia, Flux.
  
  

Data Analyst Intern @ ESA (Sep 2020 - Feb 2021)
Main duties:

• Analyse website data on traffic patterns, behaviour, navigation and user flows
• Answer key questions through statistical analyses, reporting and dashboards using analytics toolset
• Share data views to enable responsible staff to optimize decision making
• Analyse internet trends to identify future technology needs and internet patterns
• Identify methodologies and technologies for data and content linking
• Advise on Search Engine Optimisation and best practice using the Search Engine Console tool
• Assist in the review of the ESA Earth Observation Web Development Guidelines
  
  

Hackathon 2019 - Organizer @ Tor Vergata (Feb 2019 - May 2019)

• Website co-creator
• Logistic
• Development of competition themes
  
  

Professional Course Java EE @ Eustema S.p.A (Feb 2019 - Mar 2019)

• Java SE
• Java for Web with Servlet and Spring
• Database SQL and NoSQL
• Microservices
  
  

Contributing Writer @ Toward Data Science (Present)
Online publications that provides insights into machine learning and deep learning solutions.

• Selected articles:
  • Linear Regression and Gradient Descent Using Only Numpy
  • Transfer Learning for Robust Image Classification
  • Iterative Pruning Methods for Artificial Neural Networks in Julia
  • Remaking the “Silicon Valley” SeeFood App
    
    

Member @ Space Generation Advisory Council (July 2022 - Present)
The Space Generation Advisory Council is a global non-governmental, non-profit (US 501(c)3) organisation and network which aims to represent university students and young space professionals to the United Nations, space agencies, industry, and academia.

Dock3 - The Startup Lab @ Dock3 (Mar 2021 - Jun 2021)
Dock3Sprint is the incubation program of RomaTre University, where the teams go from idea to market. Each team go through 15 workshops to explore the main themes for the development of a startup: fundraising, growth hacking, lean metrics, lean management, legals, hardware production and much more. The co-working space is continually visited by mentors, industry experts, entrepreneurs and startup founder, who can monitor and support the development of the startups.- Website co-creator

Hackathon 2019 - Organizer @ Tor Vergata (Feb 2019 - May 2019)

• Website co-creator
• Logistic
• Development of competition themes
  
  

Kids Entertainer @ Isttuto S.Maria Mazzarello (Summer 2017)

Best System Award @ EVALITA 2020
Development of a stance detector system for Italian tweets exploiting using BERT with a transfer learning approach. You can read the paper here.

Hackathon winner @ ConfCooperative Hackathon (Nov 2019)
The github repo of the project. Health Cooperation application that allows user identification by eliminating paperwork, a review system of affiliated facilities and user profiling using data in order to seek solutions to provide tailored services.

Italian: Native
English: C1
Hungarian: B2

Pi School - School of AI, School of Artificial Intelligence
Pi School - Rome, Italy (May 2022)

Deep Learning Advanced Master, Deep Learning Italia
Deep Learning Italia Academy - Experis Academy - Rome, Italy (Present)

Master in Computer Science, University of Tor Vergata, Rome Italy
Thesis : "An Assessment of Iterative Pruning Methods for Artificial Neural Networks in Julia"

Advisor: Prof. Andrea Clementi

Co-Advisor: Prof. Emanuele Natale

Beachelor in Computer Science , University of Tor Vergata, Rome Italy
Thesis: "Dynamic of Bitcoin Network: empirical analysis of a full-node's neighborhood"

Advisor: Prof. Francesco Pasquale

Available here