Raphael Fontaine

The goal of my mission at Orange is to carry out a migration from the Cloudera platform to Google Cloud Platform, and implement new use cases. ➤ Designing and managing data pipelines on GCP : - Data transformation & modeling : implementation of Bronze, Silver and Gold data layers using DBT to ensure clean, structured, and analytics-ready data. - Access & Infrastructure Management : usage of Terraform to define and automate secure access controls and infrastructure as code. - Workflow & orchestration : usage of Airflow to schedule and orchestrate workflow through python scripts. - Continuous Deployment & Automation: Implementing CI/CD pipelines with GitLab to automate testing, deployment, and ensure smooth integration of changes. ➤ Migration of existing use case : - Transitioning PySpark scripts to Dataproc Serverless. - Converting SQL scripts to DBT for better modularity and maintainability. - Enhancing data validation with Google’s Data Validation Tool to ensure data integrity and quality.

The main goal of my mission at Groupe Sud Ouest is to implement a robust data platform on GCP. ➤ Designing and orchestrating data pipelines on GCP : - Developed Python scripts for extracting data from various sources (APIs, databases, files, etc.) - Implemented ETL/ELT processes to load data into Cloud Storage and ingest it into BigQuery - Fully orchestrated data pipelines using Cloud Composer (Airflow), ensuring automation and synchronization of all processing steps ➤ Data modeling and analytics optimization : - Leveraged Dataform to transform, aggregate, and model data in BigQuery, creating data marts tailored to business needs - Optimized data pipelines to ensure high performance and robustness for large-scale data processing - Monitored and maintained orchestration workflows to guarantee continuous quality and reliability of analytical processes ➤ Provision of QlikSense (or Data Studio/Looker Studio) reporting tool for business use: - Design of the data model specific to the dashboard - Data loading via SQL queries - Dashboard development, followed by delivery to the business - Automation of report

➤ Design in Python of a tool to apply configurations on Elasticsearch/Kibana in an automated way : - Deployment and configuration of an Elasticsearch/Kibana cluster: starting services, managing ports, user rights, etc. - Deployment of configurations on Elasticsearch: index templates, component templates, pipelines, etc. - Deployment of configurations on Kibana: visualizations, dashboards. ➤ Management of a non-relational Elasticsearch database : - Data ingestion, organized storage via indexes - Data processing with Elasticsearch pipelines - Field formatting with component templates - Data retention management with ilm - Cluster node management based on the amount of data ➤ Creation and optimization of data visualization tools (dashboards, visualizations) on Kibana : - Creation of visualizations/dashboards of all types (pie chart, vertical bar, data table, etc.) - Optimization of dashboards to improve loading times (data aggregation with the rollup tool)

➤ Processing in Python of carbon emission data as part of a project for COP26 (bash, netCDF, etc.). ➤ Data visualization through a web interface (JavaScript, Leaflet, ChartJS, PostgreSQL). ➤ Working in a cloud environment: creating virtual machines, disk extension, resource management according to needs (Docker, Kubernetes).

➤ Design of a front-end website using TypeScript with the React framework.