Adel Bouzidi

• Set up a Kubernetes cluster on CentOS 9 (Docker, Calico, Helm, MetalLB) with shared NFS storage. • Designed and implemented a text auto‑completion system based on an LSTM recurrent neural network developed in Python. • Trained the LSTM model using two approaches: sequential and distributed (Horovod + MPIJob Kubeflow). • Deployed the trained models (sequential & distributed) on Kubernetes. • Monitored performance and prediction accuracy via Prometheus & Grafana. • Developed and deployed an interactive web interface on Kubernetes for user demonstration. • Conducted a comparative analysis of training times and resource consumption between the two modes.

• Developed and customized SharePoint portals (on‑premise and Online) aligned with business processes. • Collaborated with functional teams to design, deploy, and maintain efficient, scalable SharePoint solutions. • Technologies: SharePoint, SharePoint Designer, SPFx, C#, JavaScript, .NET Framework, Microsoft SQL Server, Active Directory, DNS, IIS, CSS, SharePoint Server.

This self‑initiated web application showcases my personal learning in modern web development. Built with SQL Server 2016, ASP.NET Core Web API, Angular, and Bootstrap, the project’s full source code is available on GitHub.

• Set up a Kubernetes cluster (Ubuntu 20.04, Docker, Calico, MetalLB) with NFS storage. • Designed and implemented a resource‑consumption prediction system for a microservice in Kubernetes, based on an LSTM Encoder–Decoder model. • Built the training dataset: generated CPU load and collected metrics (using top) into the InfluxDB time‑series database. • Implemented predictive autoscaling driven by the LSTM, proactively adjusting a microservice’s resource allocation (vertical scaling) and the number of replicas (horizontal scaling). • Deployed and configured Grafana on Kubernetes to visualize predictions. • Deployed on Kubernetes all modules: data generation, training, prediction, visualization, and autoscaling. • Demo video: available in the link bellow.

• Designed and implemented a web application for managing the assignment of final‑year projects (PFE) at the University of Bouira. • Performed analysis and design using UML (use‑case, activity, sequence diagrams, logical data model). • Implemented the solution with the Model–View–Controller (MVC) pattern, leveraging ASP.NET / C#, JavaScript, AngularJS, Bootstrap, and Microsoft SQL Server 2012.