I build AI agents and production infrastructure for industrial companies. 10+ years bridging the physical world — sensors, machines, factory floors — with intelligent automation that runs unsupervised.

I take on consulting projects when the problem is interesting. If you're figuring out how AI agents, predictive maintenance, or infrastructure automation could work for your business, let's talk.

AI Agent Systems — Multi-agent platforms with MCP servers, cognitive memory, safety boundaries, and real-time infrastructure monitoring. Production systems, not demos.

Infrastructure — Kubernetes on bare metal, GitOps with ArgoCD, CI/CD pipelines, GPU passthrough for LLM inference, automated everything.

IIoT & Predictive Maintenance — Vibration analysis, sensor architectures, condition monitoring pipelines, and the analytics that keep machines running.

PythonTypeScriptGoKubernetesArgoCDOpenTofuProxmoxPrometheusGrafanaPostgreSQLQdrantFastMCP

• Backing Up Qdrant Snapshots Correctly: From emptyDir to NFS Persistent Backups (2026-07-29)
• Silent Drift: Why Re-Embedding Only on Count Changes Rots Your Semantic Index (2026-07-15)
• Your Vector DB Snapshots Are Landing on the Same Disk That Will Fail (2026-07-03)
• Eviction Without Deletion: Running an ACT-R Decay Policy for Agent Memory (2026-07-01)
• SealedSecrets: Storing Secrets in Git Without the Risk (2026-06-27)

_{The best technology work happens at the boundary between domains.}