Julien Castrie

o Objective: AGL sought to evaluate Microsoft Azure IoT Hub through a POC to centralize and manage data streams from connected IoT devices o Technical environment: Azure IoT Hub, Azure Storage Account, Azure Functions, Azure Event Grid, Azure Stream Analytics, Power BI, Visual Studio Code Results: - Designed and documented a POC on Azure, integrating vehicle tracking data streams - Standardized data formats and implemented automated alerts (geofencing and signal loss)

o Responsible for designing the IT architecture during the renovation of AXA's global headquarters (3 buildings, 20,000 square meters, 3,000 daily visitors) o Designed and deployed the Operational Technology (OT) network, including: CCTV systems, access control, Physical Security Information Management (PSIM), and Building Management Systems (BMS) o Contributed to the design of the IT network, covering enterprise Wi-Fi, meeting room systems, Integrated Workplace Management System (IWMS), the building’s mobile application, and the audiovisual network o Served as a jury member for RFP presentations o Technical environment: AD DS, PKI, EAP-TLS, VMware, Hyper-V, Capsule OS, Veeam Backup, Cisco switches (NAC), Palo Alto Firewalls, Evolynx, Milestone, EcoStruxure, Modbus Results: - Design phase: Gathering requirements (functional and cybersecurity) and documenting the enterprise architecture at the network level for all site applications - Implementation phase: Supervising a dozen contractors in the deployment of OT applications on segmented networks in compliance with cybersecurity requirements

o Designed a 15-day sales forecasting model incorporating weather data (time series) by location and product type, completed as a thesis project for an evening engineering program o Technical environment: Google Colab, PyTorch, Pandas, Scikit-Learn, Matplotlib Results: - Consolidated and cleaned weather and sales time series data, resulting in an optimized supervised learning dataset - Designed and trained an attention-based autoencoder model, inspired by natural language processing techniques

o Key projects: Agile management of an IoT sensor monitoring solution (+120,000 LoRaWan sensors, €2M budget) and migration of an application to a data model compliant with enterprise architecture o Coordinated business, functional, technical, and security requirements, with CAPEX/OPEX tracking o Technical environment: AWS (Lambda, Cognito, EC2…), Kafka, Oracle, Angular, Kibana, LoRaWan Results: - Delivered on time and within budget, optimized the data model, and ensured effective alignment between business and technical teams

o Full-stack development of a customer loyalty solution for independent merchants, including commercial prospecting o Technical environment: Ionic 2+, Symfony 3, Ubuntu, Apache2, RabbitMQ, iOS, Android, PostgreSQL, TLS, Jasmine, PHPUnit Results: - Implemented a robust multi-platform solution to enhance customer engagement for merchants

o Architecture: Designed architecture and sequence diagrams (PlantUML) for a connected services middleware o International Collaboration: Worked with functional, development, infrastructure, cybersecurity, integration, and end-to-end testing teams (Japan, USA) o Major Client: Nissan North America — conducted functional workshops and on-site integration tests for annual production exceeding 500,000 vehicles o Technical Environment: Azure, PCF, RabbitMQ, PostgreSQL, Redis, Java Spring Boot, OAuth2, JWT, Postman, MQTT Results: - Developed a clear architecture to facilitate communication between international teams and coordinate development efforts - Contributed to critical integration testing of a high-volume industrial system

o Solution for managing targeted customer offers and multichannel promotion execution (point-of-sale, e-commerce) o Led Maxxing projects for Metro Inc. (550 stores in Canada) and Simply Market (350 stores in France) o Key responsibilities: Pre-sales, activity and resource planning, team management (3–5 people), drafting and validating functional and technical specifications o Technical environment: RedHat, PostgreSQL, SoapUI Results: - Successfully implemented the multichannel solution, rigorously managing teams and meeting deadlines

o Pharmaceutical Industry (2 months) – Implemented a centralized accounting system for stock ownership. - Results: Improved traceability and compliance across business and IT processes o Energy Industry (5 months) – Built a spend database from SAP MM data and supported sourcing strategy. - Results: Enabled data-driven procurement decisions and supplier optimization o Energy Industry (2 months) – Structured Bills of Materials for spare parts management. - Results: Streamlined material tracking and warehouse planning o Energy Industry (3 months) – Designed KPI dashboards for project monitoring. - Results: Provided real-time visibility to stakeholders for better decision-making o Chemicals Industry (4 months) – Defined optimal stock levels to reduce costs. - Results: Lowered inventory holding costs while maintaining service levels o Imaging Industry (1 month) – Cleared SAP FI/CO invoice backlog. - Results: Restored operational stability and improved reporting accuracy

o Reverse Logistics Project Manager (3 months) – Managed all projects related to returns from stores to suppliers via recycling centers (pallets, crates, bottles, recycled materials). Oversaw financial reporting for reverse logistics costs and revenues, and led the rollout of large-scale operational procedures - Results: Reduced waste management costs and standardized reverse logistics processes o Transport Project Manager (12 months) – Optimized transport utilization and developed forecasting tools through SAP gap analysis - Results: Improved transport efficiency and reduced operational costs o Junior Project Manager (19 months) – Led warehouse relocation projects, implemented TMW logistics software, and conducted continuous supply chain optimization analyses - Results: Increased operational capacity and streamlined warehouse operations