Amine Tayari

• Design and develop scalable ETL pipelines using AWS Glue with Spark Scala, ensuring efficient data processing across large datasets. • Architect, provision, and maintain AWS infrastructure resources (S3, Glue, Lambda, IAM, Step Functions, etc.) with a focus on reliability and cost optimization. • Build and manage CI/CD pipelines to automate deployments and enforce quality gates. • Review code contributions across the team to ensure adherence to engineering best practices, data standards, and architectural guidelines. • Implement data quality checks and monitoring using AWS Glue and custom validation scripts to ensure data integrity. • Optimize Spark jobs for performance and cost-efficiency, including partitioning strategies, caching, and broadcast joins. • Define and enforce data governance policies, including access controls, encryption, and lineage tracking across the data platform. • Collaborate with data scientists and business analysts to transform raw data into analytics- ready datasets. • Publish and maintain interactive dashboards in Amazon QuickSight, enabling stakeholders to access real-time business insights securely.

• Implementation of secure infrastructures on Databricks AWS. • Development of specific tools for services, analysis, exploration, dashboard cockpits, and predictive uses. • Establishment of a transversal data heritage, governed according to Group rules, cataloged, and available for each specific use. (Glue and Collate) • Recommend best practices for development, testing and deployment. • Fine management of access to work and production environments, and to the data. • Development of a data ingestion tool in Spark/Scala with its CI/CD pipeline. • Creation of a Databricks job creation tool via Terraform. • Support for use cases in terms of technical architecture and build phase. • Development of a toolkit that will evolve with the technical and functional requirements of applications, with packages of code, libraries, available to developers.

• Migration of production applications from HDP to Cloudera. • Creation of different environments and adaptation of appropriate configurations. • Creation of CI/CD pipelines. • Creation of Terraform jobs allowing the sizing and provisioning of VMs linked to different environments. • Support for teams during the migration phase. • Development and deployment of Spark Scala libraries. • Orchestration and monitoring of jobs in production. • Design and implementation of Nifi pipelines for ingesting data exposed by external APIs to the Lake.

• Technical study of the project architecture. • Preparation of data coming from various sources so that it can be used by both applications, transformation of csv files into parquets relating to certain sources, and use of Hive tables for other entries. • Creation of Jenkins jobs to launch the different stages of the project's life cycle. • Development of a solution based on Ansible playbooks to allow developers to launch workflows in parallel. • Creation of an Oozie bundle that encompasses all the actions implementing the business rules requested in Spark Scala. • Creation of the necessary schemas for data insertion into Cassandra tables. • Insertion of calculated results after comparison with data already stored in Cassandra tables. • Solr indexing of some Cassandra tables to ensure the search functionality of the application. • Assistance to the responsible team in production deployment and monitoring the status of the application after deployment.

Study and implementation of a tree-based Machine Learning model for the detection and classification of points of interest: - Pre-processing of spatio-temporal data, definition of concepts related to the problem, selection of the model and its parameters, modification of some performance metrics and benchmarking of different classification models. - Realization of some visualization tools (interactive maps and a video) illustrating the results obtained during the internship. - Deployment of the designed project. Developed skills : Project management, Data Modeling, Interpretation of Machine Learning models applied to real situations. Used softwares : Python, R.

Study, conception and realisation of dashboard which is used to monitor operating committees of BIAT agencies. Developed skills : Project management, DATA mining, Reporting, Data modeling, Data visualization. Used softwares : QlikView, Excel, Notepad++, Java

Study of the production process of telecommunication cables and realisation of an electrical cabinet for an automatic reel. Developed skills : Project management, Team work, Ponctuality.