Benjamin D.

I run Kubernetes for teams shipping AI agents and LLM-powered products, so your agents stay up, scale with demand, and don't burn through your cloud budget. Top Rated Plus, $100K+ earned, 7+ years in DevOps and SRE. Getting an AI agent from demo to production is mostly an infrastructure problem: long-running and stateful workloads, bursty GPU and API usage, tool calls hitting internal services, and failures that are hard to see. That's the layer I own. What I do: • Kubernetes platforms: EKS, OpenShift and other managed clusters. Setup, migrations, autoscaling, upgrades, and dev/staging/prod environments • AI agent & LLM deployment: containerizing agents and model servers, GPU node pools, queues and workers for long-running agent tasks, secrets and API-key management • Observability for agents: Prometheus and Grafana dashboards for latency, errors, queue depth and API spend, with alerts that actually matter • CI/CD for AI apps: GitHub Actions, GitLab CI and Jenkins pipelines with safe rollouts and quick rollbacks for agent and model updates • Infrastructure as Code: Terraform and Ansible on AWS, GCP and Azure, so every environment is reproducible • Security & reliability: network policies, least-privilege access for agent tools, SLOs/SLIs and incident response Recent work includes long-term DevOps engagements for Blocksi and Education Ecosystem, running production infrastructure for their platforms. How I work: • I start with a short audit so we agree on priorities before I change anything • Everything is in version control and documented, so you're never locked in to me • Clear async updates, with working hours that overlap Europe and the US East Coast (UTC+1) Good fit: startups taking an AI agent from prototype to production, and teams whose Kubernetes setup is slowing them down. Message me with where you are now and I'll reply with how I'd approach it.