guillaume fradet profile picture

Guillaume Fradet

Data Scientist / Deep Learner

Deep Learning researcher at AZmed, I am fascinated by AI, Deep Learning and Computer Vision. I truly believe in the #AIforGood movement, that aims to place the AI at the service of the human and the common good. In the years to come, I aspire to join major projects in e-health, that will revolutionize the current practices.


Domain Adaptation

The key to deploy algorithms on a large scale

A powerful detection algorithm for one medical center, but unsuitable for another. This is one of the major problems faced by research teams wishing to deploy their models on a large scale. In this article, I introduce this challenge called Domain Adaptation in the context of medical imaging, based on my experience at AZmed.

Read More

Predict lipomatous soft tissue tumors malignancy on MRI

Radiomics vs. Deep Learning

Research report comparing two main approaches - Radiomics and Deep Learning - to detect malignant soft tissue tumors directly on magnetic resonance imaging.

Read More

  • Guillaume Fradet
  • July 2019

MRI brain tumor dataset

Data downloading & loading

Utilities to download (CLI) and load an MRI brain tumor dataset with Python, providing 2D slices, tumor masks and tumor classes.

Read More

  • Guillaume Fradet
  • December 2018

Computer Vision: Mars craters

Object detection / segmentation

Application of object detection / segmentation (MaskRCNN and RetinaNet) on Mars satellite images. The goal was to detect automatically craters.

Read More

Reinforcement Learning

Q-Learning implementation

Implementation of the reinforcement learning technique Q-Learning. The environment was MountainCar (OpenAI Gym), where the goal is to teach a car to swing from left to right so that it can climb the mountain.

Read More

Data viz: VAST Challenge

Visualization and Machine Learning

Development of various visualizations to respond to the VAST mini-challenge 1 of 2018. The goal was to discover if the actions of a company had led to the migration of birds. Spatio-temporal data analysis led to the development of an interactive map to observe the movements of different species. Spectral analysis of bird songs has also revealed interesting facts.

Read More

  • Guillaume Fradet
  • Dec. 2018

MAP estimation on NMF

Graphical models

Maximum a posteriori (MAP) estimation on NMF (non-negative matrix factorization), applied on faces.

Read More

A.S.T.R. - Archiving System Truly RESTful

Development of an Open Source web application - SoftBank Robotics

Development of an Open Source archiving system allowing the storage of any type of data. The users can interact with web-based system through the website or directly from scripts (using the developed Python library). Originally, the application was designed and developed within SoftBank Robotics to store tests performed on NAO and Pepper robots.

Read More

French Cup of Robotics

2016 and 2017 editions

Participation in two editions of the French Robotics Cup with DaVinciBot, association of the Pôle Universitaire Léonard de Vinci.

Read More

  • Guillaume Fradet
  • February 2018

Top Chef

Scrapping & React.js web application

A web application that makes it easy to find all Michelin starred restaurants that currently offer promotions in France.

Read More

MongoDB User Interface & API

NoSQL Databases

Web application to easily explore a MongoDB database, containing the 5000 best selling movies on Amazon.

Read More

  • Guillaume Fradet
  • March 2018

YouTube 0.1

Development of an Android application

The goal of this project was to develop an Android application similar to Youtube, using their API.

Read More


  • polytechnique

    2018 - 2019

    École Polytechnique - Master 2 Data Science

    Classes followed during the 1st semester:

    • Optimization for Data science
    • Reinforcement learning
    • Introduction to Graphical Models
    • Big Data Framework (Hadoop, Spark, ES)
    • Systems for Big Data Analytics
    • Visualization and Visual Analytics for Data Science
    • Big Data camp

    Classes followed during the 2nd semester:

    • Deep Learning
    • Advanced Learning for Text and Graph Data (NLP)
    • Data Stream Processing
    • Machine Learning, Business case
    • Big data & insurance
    • Data infrastructure (NoSQL)
    • Mixed Effects Models for the Population Approach

  • esilv

    2014 - 2019

    ESILV (Engineering school), Paris - La Défense

    Major: Computer Science, Big Data and connected objects

    Activities and associations : Member and Communication manager of the association of robotics DaVinciBot.

    Example of the classes followed:

    • Machine Learning and clustering
    • Mobile application development
    • Web Application Architectures
    • Adv topics in NoSql databases
    • Embedded Systems: architecture & programming
    • MVC: interfaces and data
    • Inferential statistics
    • Numerical analysis
    • SQL Database
    • Signal processing

  • louisiana-techn-university

    2017 - Fall semester

    Louisiana Tech University, USA

    Major: Computer Science
    Thanks to an exchange program, I had the opportunity to study in this american university for one semester.

    Article about my experience (by ESILV)

    Classes followed:

    • Adv data structures & algorithms
    • Software design & engineering
    • Computer Networks

    GPA: 4.0 (A)

  • saint-dominique

    2011 - 2014

    Saint Dominique High School, Mortefontaine

    Scientific Baccalaureat, with high honours


  • altran

    November 2019 - Present

    Data Scientist (Deep Learning Researcher) - AZmed, Paris

    AZmed automatically detects fractures on X-rays for doctors to spend more time on life-threatening exams.

    Medium article: Domain Adaptation: the key to deploy algorithms on a large scale
  • altran

    April - September 2019

    Data Scientist in Medical imaging - Altran Research, Vélizy-Villacoublay

    Application of machine learning methods on radiomic data, and deep learning techniques on MRI images, to classify, detect and segment benign and malignant tumors (lipomas and liposarcomas). Nowadays, the diagnostic of the latter is made by biopsy after an MRI scan. However, the malignant/benign ratio is very low for these tumors, so most of the biopsies are unnecessary. A diagnosis made directly from MRI images would be an interesting alternative, both for patients who would avoid an invasive examination, and for medical structures that would save money and time.
    The goal of the internship is to write a research article where different approaches and techniques are compared.

    Work in collaboration with the biomedical imaging research laboratory CREATIS.

    Report: Radiomics vs. Deep Learning to predict lipomatous soft tissue tumors malignancy on Magnetic Resonance Imaging
  • softbank-robotics

    April - August 2018

    Full Stack Developer - SoftBank Robotics Europe, Paris

    I joined the System Integration & Verification department. The members of this team are in charge of planning and performing hardware tests, in order to check the validity of the robots before their production. The internship mission was to create an archiving system to store the data from the tests performed on NAO and Pepper.
    I have developed a web application based on a RESTful API, which is now deployed and used in several departments of SoftBank Robotics.
    The application, named A.S.T.R. - Archiving System Truly RESTful, and the Python library are Open Source and available on the GitHub of the company.

    ➢ Elaboration of specifications.
    ➢ Technology choices.
    ➢ Development of the web application (API and website).
    ➢ Development of a Python library.
    ➢ Managing the sending of large files to the server.
    ➢ User management.
    ➢ Advanced archive search.
    ➢ Deployment management.

    ✹ Techs used : Node.js, MongoDB (NoSQL), Express, Python, Git, Docker, Ansible
    ✹ Environment : Linux
    ✹ Source code : the web application and the Python library

    Article: Un stage chez SoftBank Robotics : Guillaume, élève-ingénieur promo 2019, développeur Full Stack (ESILV)
  • fishfriender

    June - July 2017

    Developer Node.js - Fishfriender, Paris

    FishFriender is an online platform that aims to modernize the angling industry.

    ➢ Mission: automate data recovery and process partner data to adapt and insert them into the application database. The partner data were either in Excel sheets or in web pages. I extracted the data using web scraping techniques.

    ✹ Tech used: Node.js, JavaScript, SQL, JSON.
    ✹ Environment: Mac OS

  • davincibot

    September 2016 - June 2017

    Communication manager - DaVinciBot, Paris

    DaVinciBot is an association of Leonard De Vinci University, which aims to gather all the students interested in robotics. It was first created to participate to the French Cup of Robotics. Now we work on several projects like InMoov, a life size humanoid robot we are printing in 3D.

  • marmon

    June - September 2016

    Sales assistant - Marmon Sports, Aubervilliers

    Summer Job in a sports shop.

  • asterix

    July 2015

    Operator - Parc Astérix, Plailly (60)

    Summer Job in an amusement park.



  • Python # Numpy, Pandas, Scikit-Learn, TensorFlow, Keras, Matplotlib
  • Web development # HTML, CSS, JavaScript, jQuery, Node.js, Express.js
  • Databases # SQL (MySQL, PostgreSQL), NoSQL (MongoDB), ElasticStack
  • C# / Java
  • R

Software / Tools

  • Git
  • Jupyter
  • LaTeX
  • Unix
  • RStudio
  • VS Code
  • Android Studio
  • Adobe Photoshop


  • 🇫🇷 French # native
  • 🇬🇧 English # fluent
  • 🇪🇸 Spanish # notions


Contact me

fr en