Freelancer profile translated to English.
Back to original languageAbout Julien
Professionally reconverted as a Data Scientist in March 2020, I am highly motivated to stay up-to-date with today's technological developments. This is one of the reasons that led me to this profession, which requires us to evolve with the world, as well as my interest in development.
Communication is one of my strengths, I am attentive and proactive. In any case, I strive to move projects forward in the right direction.
My areas of activity?
Data Science: I am a graduate Data Scientist Engineer (Level 7 Diploma)
Agility: I am a certified Scrum Master (PSM I Scrum.org Certification)
Before that?
I have 15 years of experience working for major industrial groups such as Dassault Aviation and PSA. All these years have allowed me to acquire solid skills in project management, teamwork, and SCRUM methodologies. For seven years at PSA, I was responsible for creating and presenting summaries of technical work to various national and international client services. I am someone who likes to communicate, honest and open.
Personality
Autonomous: My retraining course was 100% online, which allowed me to find the necessary resources to properly assimilate many complex processes.
Curious: I continue to learn through numerous resources (Udemy - Coursera - Social Networks) and conduct daily monitoring.
To complete this presentation, I invite you to contact me and I will be happy to answer you.
See you soon.
Julien Di Giulio
English
Fluent
French
Native or bilingual
Italian
Conversational
Can work on-site
Mennecy (up to 50km), Paris (up to km), Évry (up to km), Mennecy (up to 30km)
Experience
- LeihiaData ScientistJuly 2020 - July 2020Rueil-Malmaison, Île-de-France, FranceSetting up an artificial intelligence algorithm to match the ideal candidate for the right job in the right company.
- Projet d'étudesData Scientist - AWSJanuary 2020 - March 2020 (2 months)Mennecy, Île-de-France, FranceDeploy a model in the Cloud Preparing an image classification model taking into account a future scaling of calculations due to the increase in data volume. Tools used: - Language: Python 3.6 - Environment: Linux Mint – VirtualEnv – JupyterLab – AWS – Git & GitHub - Libraries: Spark 2.4.5 – OpenCV – Boto3 Work carried out: - Code: - Preprocessing of a classifier model by loading images into a Spark dataframe. - Extraction of descriptors from each image with the ORB algorithm included in OpenCV. - Saving the final dataframe in parquet format. - Deployment on AWS: - Creation of an S3 repository, with and without AWS CLI. - Creation of an EMR instance for the distribution of calculations. - Configuration of roles in AWS IAM. - Use of an EMR Notebook to load and visualize the results.
- Projet d'étudesData Scientist - AzureDecember 2019 - January 2020 (1 month)Mennecy, Île-de-France, FranceImplement a Scoring model Predicting the risk of bankruptcy of a client for a credit company. Tools used: - Language: Python 3.7.5 - Environment: Windows – VirtualEnv – JupyterLab – Microsoft Azure – Git & GitHub - Libraries: Pandas – Numpy – Matplotlib – Seaborn – Scikit-Learn – imblearn – XGBoost – Streamlit – Flask – Docker Work carried out: - Binary classification on an unbalanced dataset. - Data preprocessing with data rebalancing. - Testing of several models: Logistic Regression – Random Forest – Xgboost. - Optimization of the most promising model. - Studies of different metrics: Accuracy – Precision – Recall – F1Score - Confusion matrix – ROC Curve – AUC Score – Features Importances - Creation of a Dashboard with Flask and Streamlit. - Deployment of the model on Microsoft Azure: Docker – Registry – Azure Service Plan – Web App Service – Azure CLI - See my LinkedIn article explaining the procedure in detail
Recommendations
Be the first to recommend Julien
Help this freelancer shine by sharing your experience working together.
These freelancer profiles also match your criteria
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Education
- Data ScientistOpenClassRooms2020Parcours Data Scientist sur la plateforme OpenClassRooms : https://openclassrooms.com/fr/paths/164-data-scientist Ce parcours est constitué de sept projets professionnalisant dont vous pouvez voir le détail dans le lien ci-dessus. Chacun de ces projets contient sa spécificité, ce qui m'a permis d'acquérir des compétences dans : - Python - Jupyter Notebook - Environnements virtuels - Le Data Mining ( Analyse, Compréhension, Nettoyage, Mise en forme, Visualisation des données) - La régression supervisée, non supervisée - La classification supervisée, non supervisée - Le clustering - Le traitement de texte et d'image - Le développement d'API - La création de Dashboard - Le déploiement sur le Cloud - Le Big Data
- Data ScientistOpenClassrooms2020Huit projets professionnalisant à valider auprès d'un jury : - Analyse exploratoire de base de données. Aide décisionnelle d’expansion à l’international d’une entreprise de cours en ligne. - Proposition d’application en lien avec l’alimentation. - Prédiction de conso de CO2 et conso totale des bâtiments de Seattle en se basant sur des bases de données existantes. - Comprendre les différents types d’utilisateurs grâce à leur comportement et à leurs données personnelles. Evaluer la fréquence à laquelle la segmentation doit être mise à jour. - Réaliser une première étude de faisabilité d'un moteur de classification en se basant sur une image et une description pour l’automatisation de l’attribution de la catégorie de l’article. - Prédiction de risque de faillite d’un client pour une entreprise de crédits. - Préparation d’un modèle de classification d’images en tenant compte d’une future mise à l’échelle des calculs due à l’augmentation du volume de données.
Certifications
- SQL for Data Science from A to Z (analysis of real data)Udemy2020