Société Générale’s Block Storage team is part of its private cloud division, delivering scalable OpenStack-based infrastructure within a SAFe Agile Release Train. Developed RESTful APIs and internal tools using Python and Flask for OpenStack-based cloud storage Deployed and maintained CI/CD pipelines with Ansible, Terraform, and Jenkins Designed tooling in collaboration with cross-functional cloud platform teams Drove roadmap execution focused on scalability and availability within SAFe ART Managed storage arrays and server patching; ensured infrastructure security compliance Monitored infrastructure using Prometheus, Thanos, and Nagios Managed over 30PB of block storage across 50+ storage arrays, fully automated for patching and security compliance Provided technical mentorship and guidance to developers Skills: Python · Kubernetes · PostgreSQL · Flask · React.js · GitHub · Git · OpenStack · Ansible · Terraform · Hashicorp Vault · Clouds privés

Jabmo is a global SaaS platform specialized in Account-Based Marketing (ABM) for industrial companies, providing analytics and behavioral targeting at scale. Built and scaled backend services with Flask and AWS (Lambda, RDS, CloudFormation, API Gateway) Delivered frontend components using React aligned with product roadmap Reduced AWS monthly infrastructure costs by $20,000 through architecture optimization and resource consolidation Maintained analytics pipeline (Snowplow) and internal tag management system Managed infrastructure improvements for performance and availability Facilitated Scrum ceremonies and mentored junior developers Deeply understood and contributed to application architecture evolution Skills: React Admin · Kubernetes · Jenkins · Docker · Python · GraphQL · SonarQube · AWS Cloud · Shell Scripting · PostgreSQL

Contributed to the design of a custom rail collision simulation engine as part of SNCF's infrastructure modernization program. Analyzed user needs and created technical specifications Developed OOP-based systems and optimized matrix calculations (NumPy, Pandas) Built a RESTful backend with Django (Master/Slave architecture) Integrated APIs with internal SNCF tools and systems Wrote end-user and technical documentation Tested using Selenium and Python Unittest framework Skills: Jenkins · Pandas · Microsoft Azure · Python · Vue.js · PostgreSQL · Django REST Framework · D3.js

Development of collision detection simulation software This software enables simulations of potential collisions between trains and obstacles or railway elements. Key contributions included: Requirements analysis: Collaborated with SNCF business teams to identify needs and define the specifications document. Calculation optimization: Enhanced calculation modules using matrix computing (NumPy, Pandas). Technical foundation: Built a robust and easily scalable technical base leveraging Object-Oriented Programming (OOP) and design patterns. Validation process management: Oversaw data validation workflows and application authorization processes. Testing and quality assurance: Conducted software component testing and acceptance (Unittest, Selenium). Documentation: Authored comprehensive documents, including: Detailed architecture descriptions. User manuals for the application. Development of a REST server application: Architecture modeling and design: Designed architecture following the Master/Slave paradigm. Implementation: Developed the application using Django. API system development: Built APIs to facilitate interactions between the software and other SNCF tools. Skills: Django REST · Jenkins · Docker · MongoDB · Pandas · Microsoft Azure · Python · Vue.js · PostgreSQL · NumPy · D3.js

BillJobs is an ERP software provider for communication agencies, supporting time tracking, invoicing, quoting, and document workflows. Migrated legacy desktop ERP to a modern web-based stack using Django, AngularJS, and PostgreSQL Designed and implemented a custom scripting language interpreter in Python to maintain backward compatibility with legacy VBScript-like logic used by 350+ client agencies Developed a dynamic PDF generation engine with ReportLab, enabling automated rendering of quotes and invoices with layout previews in PNG/JPG Built a parsing system using ElementTree to transform XML templates into visual components (headers, bodies, footers) for print-ready documents Delivered a new document modeling architecture supporting zone/group/element hierarchies with reusable interface contracts Ensured smooth client transition by replicating legacy behavior while introducing new extensibility features Implemented CI/CD pipelines using Jenkins to automate testing and deployment of new modules Authored internal documentation and supported QA via Selenium test scripts Skills: Python, Django, Flask, GIT, PostgreSQL, HTML5, CSS, JavaScript, Jenkins