Mohamed Ali Mellah
Cloud-Native & Kubernetes Engineer
Munich, Germany
Actions
I’m a Cloud-Native and Kubernetes specialist with expertise in designing, deploying, and managing Kubernetes clusters, containerization, and microservices architecture. I have hands-on experience in cloud architecture, deployment automation, and infrastructure as code, along with skills in Go programming and DevOps practices.
Area of Expertise
Topics
Your CI/CD Pipeline Is Over-Privileged: Identity Risks in Modern Cloud Platforms
Modern cloud environments depend heavily on CI/CD pipelines to deploy infrastructure, applications, and updates quickly. Over time, these pipelines often end up with far more access than they actually need, along with long-lived credentials and permissions spread across multiple cloud services.
In this talk, we look at how CI/CD systems can quietly become one of the biggest identity risks in cloud environments. Using practical examples, we go through common issues like static secrets in pipelines, overly permissive service accounts, and automation workflows with unrestricted access.
We also explore ways to reduce this risk using approaches such as short-lived credentials, identity federation, workload identity, and least-privilege access.
The goal is to show why deployment pipelines should be treated as highly privileged identities, not just automation tools, and how teams can secure them without slowing down development.
GitOps Security Anti-Patterns: When Your Repo Becomes the Breach
GitOps is widely adopted for its simplicity, auditability, and automation but it is often treated as inherently secure. In reality, many GitOps setups introduce critical security blind spots where a compromised repository, CI pipeline, or token can directly lead to a production breach.
In this talk, we will explore the most common GitOps security anti-patterns seen in real Kubernetes environments. These include secrets stored in Git, over-privileged CI/CD tokens, untrusted pull requests modifying production manifests, missing admission controls, and weak RBAC and branch protection rules.
Using concrete examples and realistic attack scenarios, I will show how these issues can be exploited in practice and how teams can fix them using policy as code, image and commit signing, admission controllers, and least-privilege access. Attendees will leave with a clear checklist to harden their GitOps workflows without sacrificing developer velocity.
How Clusters Fail Under Attack and How We Detect It at Runtime with eBPF signals
Most Kubernetes security work today happens before deployment. Teams scan images and secure their pipelines, but incidents usually happen after the container is already running. If someone gets access to a pod, many traditional tools provide very little visibility into what the workload actually does on the node.
In this session I introduce runtime security with eBPF, a Linux kernel capability that lets us observe system behavior in real time without modifying applications or containers. We will look at how real attacks appear in Kubernetes environments such as reverse shells, privilege escalation attempts, and unexpected process execution.
Using practical examples and open source tools, I will demonstrate how these activities can be detected and investigated, and how runtime detection complements preventive security controls. The goal is to give attendees a clear mental model of runtime threats and a practical starting point for adding detection to their production clusters.
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top