Moncef Bousselat

Designing and operating multi-cloud infrastructure — AWS and Azure — with Kubernetes clusters on top, across several client environments at once.

• Infrastructure as code in Terraform, and the automated CI/CD pipelines that deploy it.
• Security and compliance controls, including IAM and WAF.
• Observability and autoscaling for production workloads.

The SAA-C03 certification, covering resilient and cost-aware architecture design on AWS.

A blockchain identity system built on the Self-Sovereign Identity model, using Hyperledger Indy — identity data belongs to the person it describes rather than to whichever service is asking.

The Aca-Py agents that act on the blockchain run as Docker containers on ECS Fargate. The application tiers around them — a Vue front end and a FastAPI and MongoDB back end — run on EC2, deliberately split across public and private subnets so only the front end is reachable from outside.

A second Master’s degree, funded by an Erasmus Mundus scholarship and taught across three countries — a semester in France, a semester in the United Kingdom, and a semester in Sweden — with the final months spent on the thesis project in France.

The specialisation is green networking and cloud computing: building and operating distributed systems with their energy cost treated as a design constraint rather than an afterthought.

This is where most of the cloud and infrastructure work took shape, including the Terraform and Kubernetes projects.

Imported assets were going missing, and each loss cost the company up to 75% of the asset’s original value in penalties. The fix was a tracking system with a tamper-evident record of custody.

A REST API handling over 1,000 transactions a day sits in front of a private Hyperledger Fabric network, with a Flask and Vue application for the people actually logging and querying movements. The API runs on API Gateway, the blockchain on ECS Fargate, and the front and back ends on EC2.

An internship spent on how networking actually behaves in IoT systems, and where deep learning can be applied inside them to improve efficiency and security.

The concrete output was a denoising convolutional autoencoder for cleaning up noisy signal data.

Monitoring the construction materials market in Algeria: collecting the data, then analysing price trends and the quantities imported and exported over the preceding decade.

Freelance work analysing startup projects and identifying the technical complexity each was trying to innovate around — separating the genuinely hard engineering problem from the pitch surrounding it.

Research on processing, cleaning and classifying data drawn from the company’s ERP system, and on detecting outliers in it.

A short first internship — an introduction to working in a company, to web development, and to running work through an agile process with a team.

Five years at the Polytechnic School of Algiers, graduating in 2023 with a Master’s in Data Science and Artificial Intelligence.

Most of the side projects from this period came out of it one way or another — the wildfire simulation, the data visualisation work, and a steady run of machine learning experiments done largely for their own sake.