Usman Abid Butt

Client Project – Customer Satisfaction Analytics for Railway Passengers – Implemented techniques for data collection, cleansing, structuring, enrichment, aggregation, and modeling – Developed Spark jobs in Scala packaged as JAR applications – Automated deployment of Spark jobs on Databricks clusters – Exposed analytical tables through Databricks SQL Analytics Endpoint – Delivered data visualizations in Power BI via the Databricks SQL connector

Client Project – ATESS (Train "Black Box" Data) End-to-end Data Platform Development, from Data Collection to API Exposure Backend – Developed Spark applications in Scala packaged as JAR files – Implemented data ingestion and data quality pipelines following company best practices – Automated deployment of applications to Databricks clusters via CI/CD pipelines – Implemented performance optimizations including partitioning, broadcast joins, caching, and workload distribution – Used Azure cloud services such as Key Vault, Blob Storage, and Azure Table Storage Frontend (Microservices Architecture) – Developed a Python web application using the Flask framework – Implemented secure Active Directory authentication with JWT token exchange between client and server – Integrated Azure services such as Azure Table Storage and Key Vault – Generated automated API specifications using Swagger (Flask-RESTPlus) – Developed unit tests and performance tests using Locust – Built Docker images for deployment on AKS via Azure Container Registry – Automated build and deployment pipelines using Ansible and Helm – Developed an administration interface using JavaScript and HTML – Produced automated API documentation compliant with the OpenAPI Specification (OAS)

Client Project – Vegetation Monitoring: Geolocation of Incident Events on the SNCF Railway Network – Developed a Python 3 application using the Pandas framework – Implemented and improved features for geolocating incident events – Built data quality pipelines using the Luigi workflow framework – Implemented automated testing (unit tests and non-regression tests) – Produced technical documentation using Sphinx, hosted on Azure Cloud

– Development of a hybrid application (Desktop & Mobile) enabling production teams to search for spare parts using data from multiple information systems. Backend: – Elasticsearch for data indexing – Node.js application using the Express.js framework – HTTPS traffic management – Database authentication (Elasticsearch Security) – Implementation of automated deployment methods in the Cloud – Installation and maintenance of the different layers (routing, databases, applications, system, etc.) Frontend: – Application development using Apache Cordova – Testing performed with Jasmine – Production deployment and industrialization of the solution on the internal store of SNCF Additional responsibilities: – Management of different environments (development, integration, production, etc.) – Monitoring ensured through an internal solution.

– Implementation of a data catalog to map and reference datasets available in a Data Lake-type storage environment: – Import/export of large volumes of data using applications developed in Python – Handling multiple file formats (CSV, XML, PDF, images, etc.) – Performance optimization in a cloud environment using Amazon Web Services (AWS)

– Mise en place d' outils de monitoring des activités de Data M. – Les activités concernent les traitements, transformations et chargements des données massives dans les systèmes d'information de la SNCF. – Tâches effectuées : – Traitement (nettoyage, filtrage, extraction....) de données massives – Création de logs applicatifs – Développement et intégration avec ElasticSearch – Génération de Dashboard dans Kibana – Planification et déploiement d'applications – Conception et développement de solutions techniques basées sur des portefeuilles de services Cloud – Utilisation de services AWS (EC2, lambda, S3 ...) – Réalisation page Web avec statut des activités en temps réel – Intégration et automatisation de scripts python