Fabrizio Miano

Data Scientist | Machine Learning Engineer | Backend Developer

Catania, Italy

Summary

ML Engineer, Full-stack Python developer, and System administrator at R2M Solution focused on research activities within the scopes of robotics, artificial intelligence, big data, energy saving, and IoT domains. Previously Data scientist for Leonardo company involved in the development of machine learning algorithms and AI solutions applied to public administration and defence. Briefly Postdoctoral Research Fellow at University of Sussex working on deception detection, and Junior Data Scientist in collaboration with Deckchair.com working on the development of an AI algorithm capable of selecting and classifying interesting images (fireworks, sunset, sunrise, etc.) within a dataset. PhD in Experimental Particle Physics at University of Sussex/CERN and former member of the ATLAS Experiment at the LHC with extensive experience in data analysis

Languages:

English, French, Italian, Spanish

Favorite Python Packages:

Flask, pandas, spaCy, nltk, openCV, librosa, beautifulsoup, selenium

Experience

I am a ML Engineer, Full-stack Python developer, and System administrator at R2M Solution, focused on research activities within the scopes of robotics, artificial intelligence, big data, energy saving, and IoT domains. Previously, I was a Data Scientist at Leonardo Company working on a number of projects within the scopes of public administration, energy market, and defence. The projects I have been involved in required expertise on data crawling, text, audio, and image processing, as well as team management, project coordination time- and budget-wise. Previously, as a Research Fellow at the University of Sussex, I was focusing on the development of a machine learning algorithm for the classification of video of trials as truthful/deceptive.

In 2018 I graduated as a PhD in Experimental Particle Physics at the University of Sussex where I developed skills such as, problem solving,time management, coding (Python, C++, HTML, bash), and to carry out data analyses of high rates and volumes of data collected by the ATLAS Experiment at the Large Hadron Collider (CERN, Geneva). During my PhD I have been invited to give talks at national (UK), and international (US), conferences; I have also participated in outreach activities to engage with the general public.

---

IT Engineer R2M Solution Sicily, Italy

Development of internal tools using python language and front-end technologies; System administrator for internal tools and partner solutions; Technical activities in research projects within the robotics, artificial intelligence, big data, energy saving, and IoT domains; Deliver high quality reports for research and innovation projects

Jan 2020 - Present

Data Scientist Leonardo Sicily, Italy

Data crawling: development of a Python tool to scrape LinkedIn jobs and user profiles; Recommendation engine: development of a job-user recommendation system; Computer Vision: development of a WebApp to perform real-time object detection using YOLO; NLP: development of a deep-learning tool to extract the objects seized to criminal organizations from legal verdicts; Social-network Analysis: named-entity extraction and sentiment analysis on Facebook-post comments; Web application development and deployment as a support to the developed data-science models: Python (Flask, Django), HTML5, CSS, JS Dev Ops & cloud: Docker, Azure, gunicorn, NGINX;

Jan - Dec 2019

Post-Doctoral Research Fellow in data science University of Sussex Brighton, UK

Development of ML model to classify videos of trials as Truthful/Deceptive; Exploration of potential collaboration with other areas of research

Sep - Dec 2018

Junior Data Scientist Deckchair.com Brighton, UK

Development of a statistical tool to select interesting images within a dataset; Design of a ML tool to classify grey sky and sunset images

Jun - Aug 2018

---

University of Sussex / CERN Ph.D. in Experimental Particle Physics Brighton, UK / Geneva, Switzerland

• Developed Python code to perform multi-variable signal-over-background optimizations to define signal-enriched regions to enhance signal contribution while discriminating against background
• Modelling of main backgrounds using custom C++ framework
• Data-Driven estimation of an irreducible background and evaluation of related theory uncertainties (C++, Python, Excel)
• Validation of the changes to the used framework before release
• Monitoring of the performance of the tracking efficiency
• Participation in the development and commissioning of vertex reconstruction algorithm (C++)
• Development of a monitoring tool (Python) to display the results of the monitoring (plots, tables) on a web page (HTML)

2014 - 2018

Università degli Studi di Catania MSc in Physics Catania, Italy

• Developed C++ code to simulate physical distributions of the emission of protons from a nuclear source of a given size and life-time; Validation of the simulations
• Deduction of the geometrical size and life-time of a nuclear source created in a heavy-ion collision at intermediate energy using data collected with the LASSA detector of the NSCL, Michigan, US.

GPA: 107 / 110

2009 - 2014

---

Data Science & Web App Development

• Computer Vision: OpenCV, sklearn & skimage
• NLP: spaCy, NLTK
• Audio processing: Librosa
• Web crawling: Selenium, BeautifulSoup
• Data and maths: MongoDB, SQL, pandas, numpy
• Neural Networks: Keras
• Backend: Flask, Django

Teaching

• Data-science group mentor
• Classical Physics and Electromagnetism
• Properties of Matter

Languages

• Italian (Mother tongue)
• English (Fluent)
• Spanish (Fluent)
• French (Basic)

---

Public Speaking Discnet Showcase Royal Society London, UK

“FROM PARTICLE PHYSICS TO COMPUTER VISION”
10-min talk on the ML tool developed for Deckchair.com

Dec 2018

HLT-UK University of Oxford Oxford, UK

“HLT TRACKING PERFORMANCE”
15-min talk: Presented the result of the performance of the tracking of the ATLAS Inner detector

Sep 2017

Phenomenology 2017 University of Pittsburgh Pittsburgh, PA, USA

“SEARCHES FOR DIRECT PRODUCTION OF THIRD GENERATION SQUARKS WITH THE ATLAS DETECTOR”
5-min talk: Presented the results of the searches for third-generation squarks on behalf of the ATLAS Collaboration

May 2017

International Conference on High Energy Physics (ICHEP) 2016 Sheraton Hotel Chicago, IL, USA

“THE DESIGN AND PERFORMANCE OF THE ATLAS INNER DETECTOR TRIGGER FOR RUN 2 COLLISIONS AT √s = 13 TEV”
Poster presented on behalf of the ATLAS Collaboration: Results of the performance of the ATLAS Inner Detector

Aug 2016

Skills

Apache, Artificial Intelligence, Backend Development, Big Data, Data Science, DevOps, Docker, Flask, Fullstack Development, Git, Google Cloud Platform (GCP), Gunicorn, Heroku, JavaScript, Keras, Linux, Machine Learning, Matplotlib, Microsoft Azure, MongoDB, MySQL, Natural Language Processing, Nginx, NumPy, Pandas, PyTorch, Redis, SciPy, Selenium, Web Scraping