Augustin Nollevaux

Context : to comply with the European IFRS17 standard, Credendo required a new data model and ETL process to streamline reporting and operational systems for credit insurance. • I led the transition from a Power BI prototype to a robust SSIS-MS SQL Server solution, significantly enhancing performance and reliability. • I designed and industrialized ETL pipelines across four Azure cloud environments (D-Q-V-P), implementing DevOps for seamless deployment. • I improved data loading performance by a factor of 16 (reducing load times from 240 minutes to 15 minutes) • As the team’s SSIS expert, I standardized methodologies and oversaw the industrialization of multiple prototypes, including those developed by my colleagues. • I collaborated closely with operational teams in Brussels and Poland, ensuring technical solutions aligned with business needs and producing over 30 documentation files tailored for both developers and business stakeholders. • I presented weekly reports to stakeholders, including data quality checks, validations from the business, project progress, achievements, and actionable insights. Tools : SSIS, SSMS, PowerBi, Github, Devops

Context : get insights from operational data to improve the distribution network planning and design. • Designed, implemented, and analyzed dashboards to monitor KPIs related to courrier efficiency and route optimization of bicycles deliveries. • Ensured 99% data accuracy through daily quality checks on 30+ SQL tables powering 20+ dashboards, using SQL stored-procedures. • Provided actionable insights for strategic planning by segmenting 581 municipalities into five distinct groups using a Python-based ML model. Tools : PowerBI, Oracle SQL Server, Python.

Context : as part of the Client Innovation Center (CIC) at NTT Belgium, I worked on an IoT-driven project aimed at leveraging sensor data to facilitate and automate decision-making in stormwater management. The project focused on predicting water levels in a stormwater retention basin using data from optical sensors and advanced cloud-based technologies. • I developed a predictive model that predicts water levels with a 90% accuracy rate, enabling timely interventions and reducing flood risks. • I used Microsoft Azure Custom Vision and Azure Cognitive Services to process, analyze, and visualize sensor data securely stored in the cloud. Tools : Azure Cognitive Services - Custom Vision, Python, Microsoft Office.