Applied AI engineer. I build open-source reliability tools for AI agents: Argus (github.com/PRINCE2-AI/Argus), Agent Reliability OS, MCP Sentinel Lab, nano-AgentRL, and DevMind.I write The Reliable Agent, a weekly newsletter about agent evals, MCP security, and production failures: thereliableagent.substack.com