<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Scott Wueschinski — AI &amp; Agentic POV</title><description>Production POV on Claude, MCP, agent design, eval harnesses, and the agentic stack — from someone shipping it.</description><link>https://scottwueschinski.com/</link><item><title>AI governance is a logging problem before it&apos;s a policy problem</title><link>https://scottwueschinski.com/pov/ai-and-agentic/ai-governance-is-a-logging-problem/</link><guid isPermaLink="true">https://scottwueschinski.com/pov/ai-and-agentic/ai-governance-is-a-logging-problem/</guid><description>Enterprises write AI policy before they can observe what their agents do. You cannot govern what you do not log. Instrumentation comes first.</description><pubDate>Wed, 10 Jun 2026 00:00:00 GMT</pubDate><category>ai-governance</category><category>ai-governance</category><category>observability</category><category>agentic-ai</category><category>eu-ai-act</category><category>production-ai</category></item><item><title>Context engineering is the job prompt engineering pretended to be</title><link>https://scottwueschinski.com/pov/ai-and-agentic/context-engineering-is-the-job/</link><guid isPermaLink="true">https://scottwueschinski.com/pov/ai-and-agentic/context-engineering-is-the-job/</guid><description>Prompt wording barely moves the needle in production. Context assembly does. Here is what actually breaks when you treat the window as a scarce resource.</description><pubDate>Fri, 29 May 2026 00:00:00 GMT</pubDate><category>production-llm-deployment</category><category>context-engineering</category><category>ai-agents</category><category>rag</category><category>production-llm</category><category>agent-memory</category></item><item><title>Your agent doesn&apos;t need more tools. It needs fewer.</title><link>https://scottwueschinski.com/pov/ai-and-agentic/your-agent-needs-fewer-tools/</link><guid isPermaLink="true">https://scottwueschinski.com/pov/ai-and-agentic/your-agent-needs-fewer-tools/</guid><description>Tool sprawl is the silent killer of agent reliability. Cutting tools, not adding them, is what makes agents production-grade.</description><pubDate>Wed, 20 May 2026 00:00:00 GMT</pubDate><category>agent-design</category><category>agent-design</category><category>tool-use</category><category>context-engineering</category><category>reliability</category><category>mcp</category></item><item><title>Evals, or it&apos;s vibes</title><link>https://scottwueschinski.com/pov/ai-and-agentic/evals-or-its-vibes/</link><guid isPermaLink="true">https://scottwueschinski.com/pov/ai-and-agentic/evals-or-its-vibes/</guid><description>Every production AI agent needs an eval harness. Without one, you&apos;re shipping vibes — and you&apos;ll only find out when the agent acts on bad data at scale, in production, repeatedly.</description><pubDate>Wed, 29 Apr 2026 00:00:00 GMT</pubDate><category>agent-evals</category><category>AI Agents</category><category>Evals</category><category>Production Systems</category><category>CODN</category></item><item><title>MCP is the protocol the agent conversation needed</title><link>https://scottwueschinski.com/pov/ai-and-agentic/mcp-is-the-protocol-the-agent-conversation-needed/</link><guid isPermaLink="true">https://scottwueschinski.com/pov/ai-and-agentic/mcp-is-the-protocol-the-agent-conversation-needed/</guid><description>Most &apos;AI agents&apos; fail not because the model is wrong but because the agent has no way to read your data and act on your systems. MCP is the missing layer. Here&apos;s why it&apos;s the unlock.</description><pubDate>Wed, 22 Apr 2026 00:00:00 GMT</pubDate><category>mcp-and-tools</category><category>MCP</category><category>AI Agents</category><category>Claude</category><category>Production Systems</category></item><item><title>Why Claude Code is the RevOps hire of 2026</title><link>https://scottwueschinski.com/pov/ai-and-agentic/why-claude-code-is-the-revops-hire-of-2026/</link><guid isPermaLink="true">https://scottwueschinski.com/pov/ai-and-agentic/why-claude-code-is-the-revops-hire-of-2026/</guid><description>RevOps is becoming a build job. Most teams can&apos;t hire fast enough at the new bar. Claude Code closes the gap — if you treat it like a junior engineer, not a chatbot.</description><pubDate>Wed, 15 Apr 2026 00:00:00 GMT</pubDate><category>claude-engineering</category><category>AI Agents</category><category>Claude Code</category><category>RevOps</category><category>CODN</category></item></channel></rss>