28.10.2026 aktualisiert
100 % verfügbar
Principal Engineer - Platform Engineering, SRE, Cloud & AI/ML Platforms
Berlin, Deutschland
Nur Remote
M.Eng., Computer Systems Networking and Telecommunications, ENET'Com Sfax, Tunisia (2010-2012)Skills
APIsAmazon Elastic Compute CloudCloud-EngineeringCloud FoundryCloudtestingApache CloudStackMicrosoft Platform BuilderAzure Machine LearningGoogle CloudEnterprise Application PlatformRemote Application PlatformCloud PlatformOnapsis PlatformIntegration der Power PlatformIstio
Cloud Infrastructure & Platform Engineering
Expertise in designing, building, and operating cloud-native platforms across AWS, GCP, Azure, and Hetzner. Proficiency in Kubernetes (EKS, GKE, AKS), Terraform, GitOps (Argo CD, Atlantis), service mesh (Istio, Linkerd), and internal developer platforms using Backstage. Delivery of cloud-native transformations for leading enterprises including Renault, Deutsche Bank, Spryker, and Roam AI.
AI/ML Platform & MLOps
Deep knowledge in production MLOps and LLMOps on Kubernetes, including multi-LLM and multi-GPU serving with vLLM and KServe. Experience with agentic systems using LangGraph, LangChain, and MCP, as well as model lifecycle management, evaluation, and observability for AI workloads using LangFuse and OpenTelemetry.
Site Reliability Engineering (SRE) & Observability
Extensive background in SRE practices at scale, including SLO/error-budget programs, incident management, and reliability strategy. Led observability transformations from New Relic to OpenTelemetry, Prometheus, and Grafana (Loki, Tempo, Mimir), covering more than 3,000 distributed services and generating recurring enterprise revenue.
Infrastructure as Code & CI/CD
Strong command of Terraform and Terragrunt for infrastructure automation, combined with CI/CD pipeline design using GitHub Actions, GitLab CI, and Argo CD. Established GitOps-led delivery processes with immutable container images, security automation, and in-cluster smoke tests.
Python & Go Development
Implementation of backend tooling, Kubernetes operators, and control planes in Python (FastAPI, FastMCP) and Go. Development of production-grade, multi-tenant agentic services and internal APIs with reliability and security as first-class concerns.
Security & Access Management
Implementation of self-managed RBAC access, DEX-based SSO, IAM, HashiCorp Vault, External Secrets, Kyverno policies, and cert-manager. Acknowledged by the Microsoft Security Response Center for a 2020 AKS privilege-escalation disclosure.
GPU Operations & LLM Serving
Operation of mixed CPU and GPU Kubernetes clusters supporting approximately 100 self-hosted ML and language models. Experience with NVIDIA device plugins, KServe, vLLM, KubeRay, and Karpenter for scalable model serving and training pipelines.
Engineering Leadership & Mentorship
Leadership of engineering and product teams at Roam AI, AUTO1, ING, and Spryker. Management of teams of up to 10 engineers spanning MLOps, data, platform, SRE, and AI engineering, with a focus on 1-on-1s, 360-degree feedback, performance reviews, and personal development plans.
Google Cloud Certifications
Holder of Google Cloud Professional Cloud Architect (2019) and Google Cloud Associate Cloud Engineer (2018) certifications, demonstrating validated expertise in GCP architecture and engineering.
Expertise in designing, building, and operating cloud-native platforms across AWS, GCP, Azure, and Hetzner. Proficiency in Kubernetes (EKS, GKE, AKS), Terraform, GitOps (Argo CD, Atlantis), service mesh (Istio, Linkerd), and internal developer platforms using Backstage. Delivery of cloud-native transformations for leading enterprises including Renault, Deutsche Bank, Spryker, and Roam AI.
AI/ML Platform & MLOps
Deep knowledge in production MLOps and LLMOps on Kubernetes, including multi-LLM and multi-GPU serving with vLLM and KServe. Experience with agentic systems using LangGraph, LangChain, and MCP, as well as model lifecycle management, evaluation, and observability for AI workloads using LangFuse and OpenTelemetry.
Site Reliability Engineering (SRE) & Observability
Extensive background in SRE practices at scale, including SLO/error-budget programs, incident management, and reliability strategy. Led observability transformations from New Relic to OpenTelemetry, Prometheus, and Grafana (Loki, Tempo, Mimir), covering more than 3,000 distributed services and generating recurring enterprise revenue.
Infrastructure as Code & CI/CD
Strong command of Terraform and Terragrunt for infrastructure automation, combined with CI/CD pipeline design using GitHub Actions, GitLab CI, and Argo CD. Established GitOps-led delivery processes with immutable container images, security automation, and in-cluster smoke tests.
Python & Go Development
Implementation of backend tooling, Kubernetes operators, and control planes in Python (FastAPI, FastMCP) and Go. Development of production-grade, multi-tenant agentic services and internal APIs with reliability and security as first-class concerns.
Security & Access Management
Implementation of self-managed RBAC access, DEX-based SSO, IAM, HashiCorp Vault, External Secrets, Kyverno policies, and cert-manager. Acknowledged by the Microsoft Security Response Center for a 2020 AKS privilege-escalation disclosure.
GPU Operations & LLM Serving
Operation of mixed CPU and GPU Kubernetes clusters supporting approximately 100 self-hosted ML and language models. Experience with NVIDIA device plugins, KServe, vLLM, KubeRay, and Karpenter for scalable model serving and training pipelines.
Engineering Leadership & Mentorship
Leadership of engineering and product teams at Roam AI, AUTO1, ING, and Spryker. Management of teams of up to 10 engineers spanning MLOps, data, platform, SRE, and AI engineering, with a focus on 1-on-1s, 360-degree feedback, performance reviews, and personal development plans.
Google Cloud Certifications
Holder of Google Cloud Professional Cloud Architect (2019) and Google Cloud Associate Cloud Engineer (2018) certifications, demonstrating validated expertise in GCP architecture and engineering.
Erfahrung
Berufserfahrungmehr als 10 Jahre
Freelancer seit2025
Sprachen
DeutschGrundkenntnisseEnglischverhandlungssicherFranzösischMuttersprache
Projekthistorie
Own and operate the full cloud and MLOps Kubernetes platform supporting billions of daily data events, 200+ GitOps Argo applications, and approximately 100 GPU machines. Lead a team of 10 engineers. Established GitOps-led delivery, self-managed RBAC, DEX-based SSO, Vault, Kyverno policies, and Prometheus/Grafana monitoring. Established the AI software factory with repository-owned agent harnesses and deterministic gates. Led design and deployment of the Roam solution in customer-controlled environments on GCP and Azure.
Built the cloud platform from scratch on AWS and GCP with Terraform, EKS, GKE, Helm, Argo CD, and GitHub Actions. Developed production-grade, multi-tenant agentic services with remote HTTP MCP servers in Python and FastMCP. Implemented end-to-end observability for AI agents and microservices with OpenTelemetry, Grafana Cloud, Prometheus, and CloudWatch.
Architected and operated more than 300 customer cloud environments across AWS and on-premises hardware with 3,000+ distributed services. Led observability transformation from New Relic to OpenTelemetry, Grafana, Loki, Tempo, and Prometheus, saving 1.5 million dollars annually and productizing the platform for five enterprise clients. Established internal developer platform with Terraform, Terragrunt, Backstage, Argo CD, and Keptn. Designed and operated internal AI platform on EKS for multi-GPU, multi-LLM, and multi-agent workloads. Led the Cloud Platform Reliability Engineering group across three cloud product teams.