Andrea Moscatelli, Developer in Florence, Metropolitan City of Florence, Italy
Andrea is available for hire
Hire Andrea

Andrea Moscatelli

Verified Expert  in Engineering

Machine Learning and AI Developer

Location
Florence, Metropolitan City of Florence, Italy
Toptal Member Since
January 7, 2021

Andrea is a mathematician-computer scientist who graduated with 110/110 at the University of Florence and has eight years of professional work experience as a software developer, six of which were in Geneva working as a staff member at CERN - The European Organization for Nuclear Research. Andrea is skilled in solving logical problems through the application of specific algorithms, with experience in artificial intelligence and machine learning techniques.

Portfolio

Gucci
Kotlin, Amazon Web Services (AWS), IntelliJ IDEA, Java, BigQuery, Python...
Self-employed
SQL, PyCharm, Data Science, Deep Learning, Scikit-learn, Seaborn, Matplotlib...
CERN
SQL, Algorithms, R, Scrum, Linux, Eclipse, Agile, Control Systems, Physics, Java

Experience

Availability

Part-time

Preferred Environment

Eclipse, PyCharm

The most amazing...

...application I've developed can analyze a video and recognize the performed human actions with an accuracy of over 90%.

Work Experience

Full-stack Developer

2021 - PRESENT
Gucci
  • Developed the Gucci product catalogue API, used all over the world by the partners to always be updated on the brand's products.
  • Configured the needed AWS tools (ES, SQS, SNS) through Terraform scripts.
  • Analyzed the behavior of the customers on the website and the related purchases to improve the web experience recommended articles.
Technologies: Kotlin, Amazon Web Services (AWS), IntelliJ IDEA, Java, BigQuery, Python, Terraform, Datadog

Machine Learning Developer

2020 - PRESENT
Self-employed
  • Developed an application capable of analyzing the stock market (classical or cryptocurrency-related) and suggest the right moment to buy or sell specific assets.
  • Designed the entire application from data extraction to data manipulation ending to prediction model configuration.
  • Applied dozens of machine learning algorithms and statistical analysis techniques.
Technologies: SQL, PyCharm, Data Science, Deep Learning, Scikit-learn, Seaborn, Matplotlib, Cryptocurrency, Stock Market, Finance, NumPy, Artificial Neural Networks (ANN), Regression Modeling, Pandas, Algorithms, Statistics, Machine Learning, Keras, TensorFlow, Python, Artificial Intelligence (AI)

Software Developer and SPS Particle Accelerator Operator

2011 - 2017
CERN
  • Developed more than ten software applications for the CERN Control Centre, focused on helping the operators of five different particle accelerators.
  • Developed part of the logic of the "Software Interlock System," the main logic software system aimed at protecting both the personnel of operations in highly radioactive areas and the machine itself.
  • Configured and optimized the SPS particle accelerator for more than 600 days as the operator in charge and guarantor of the safety of both the personnel and the machine itself.
  • Participated in a very dynamic and multicultural environment, managing and coordinating multiple requests from teams with entirely different backgrounds.
Technologies: SQL, Algorithms, R, Scrum, Linux, Eclipse, Agile, Control Systems, Physics, Java

Software Developer

2010 - 2011
Hitech s.p.a.
  • Implemented dozens of new features to help the doctors and the staff of five different hospitals manage patient appointments and visits.
  • Tested all written code with the test framework JUnit.
  • Maintained and fixed the bugs of three different hospital products (CupWeb, AcceWeb, and PsNet) used by hundreds in Italy.
Technologies: SQL, JUnit, MySQL, Eclipse, Java

Stock Market Predictor

A Python-based application capable of suggesting the right time to buy or sell a list of specific assets in both the traditional stock market and the cryptocurrency market. I designed and implemented all the application phases, from data retrieval to configuring the predictor model.

Human Actions Recognizer

https://github.com/Andrea-moscatelli/tesiMagistrale
I developed an application capable of extracting the subjects' skeleton spot in a video and recognizing their actions with an accuracy of over 90% using artificial neural networks and deep learning techniques.

CERN Software Interlock System

The main software logic system used at CERN to guarantee the safety of the personnel and the particle accelerators. I developed part of the logic used in the SPS (Super Proton Synchrotron) accelerator and designed and transformed the safety rules into XML and Java code.

CERN Fast Extraction Interlock

A Java-based graphical interface aimed at helping the operators in the CERN Control Centre in the management and configuration of the magnets of the SPS particle accelerator extraction line. I designed and developed the entire application, implementing the best logic to help the operators in this crucial and challenging task.
2017 - 2020

Master's Degree in Data Science

University of Florence - Florence, Italy

NOVEMBER 2020 - PRESENT

Advanced Machine Learning: Deep Learning

Coursera

OCTOBER 2020 - PRESENT

Machine Learning

Coursera

Languages

Python, Java, XML, SQL, R, HTML, C++, PHP, CSS, Kotlin

Paradigms

Data Science, Parallel Programming, Agile, Scrum

Platforms

Jupyter Notebook, Eclipse, Linux, NVIDIA CUDA, Amazon Web Services (AWS)

Other

Machine Learning, Deep Learning, Algorithms, Computer Vision, Statistics, Data Mining, Control Systems, Clustering, Regression Modeling, Artificial Neural Networks (ANN), Classification Algorithms, Image Classification, Deep Neural Networks, Computer Vision Algorithms, Artificial Intelligence (AI), Bayesian Statistics, Internet of Things (IoT), Physics, Detectron2, PoseNet, Finance, Stock Market, Cryptocurrency

Frameworks

Swing, JUnit

Libraries/APIs

TensorFlow, NumPy, Keras, Scikit-learn, Pandas, Matplotlib, PyTorch, SciPy

Tools

PyCharm, Seaborn, TensorBoard, MATLAB, Weka, Git, Jira, IntelliJ IDEA, BigQuery, Terraform

Storage

MySQL, Datadog

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring