Expertise
These are the domains I've spent years mastering in production — running platforms at scale, hardening how teams ship, and now orchestrating agentic AI to take the grind out of the work. Each one is backed by real outcomes, not promises.
I design and run Kubernetes platforms and cloud infrastructure that teams can self-serve and trust in production — from EKS fleets at scale to GitOps delivery that makes deployments safe and repeatable.
Proven at 1,000+ nodes behind a 100B-events/day pipeline, with $100K+/month in cloud cost removed through platform standardization. Also ran the AWS platform behind a production RAG application.
I build agentic workflows and AI-assisted automation that take the grind out of engineering and operations — using agents, MCP, and human-in-the-loop gates to make teams measurably faster without giving up control. My focus is orchestrating AI to improve real processes, not model training or inference.
Building a gated multi-agent delivery workflow and a hybrid-memory MCP server that gives agents persistent, relationship-aware context.
I build CI/CD pipelines that make the secure path the default — rootless container builds, automated vulnerability and supply-chain scanning, and policy-as-code that stops misconfigurations before they ship.
Standardized secure pipelines that cut production vulnerabilities by 40% and build/deploy times by 80%, with rootless Docker, Trivy, SBOM generation, and Kyverno policy enforcement.
I keep production resilient and lean — designing for failure so outages don't become incidents, and finding the cloud spend that isn't earning its keep.
Recovered a 1,000+ node fleet within 2 hours of a multi-day AZ outage with zero data loss; drove $100K+/month in cloud cost savings through platform standardization and workload migration.