<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Kyler Berry — AI Engineering Consultant</title><description>Thoughts on AI engineering, automation, and building production AI systems.</description><link>https://kylerberry.com/</link><language>en-us</language><item><title>The n8n Rebuild Report: Real Numbers, Real Trade-offs, No Clean Ending.</title><link>https://kylerberry.com/blog/posts/30-days-of-n8n-heres-what-the-numbers-actually-say.md</link><guid isPermaLink="true">https://kylerberry.com/blog/posts/30-days-of-n8n-heres-what-the-numbers-actually-say.md</guid><description>I predicted $44-70/month. It came in at $90. Here&apos;s the honest breakdown - cost, stability, creative quality, and the trade-offs that don&apos;t clean up nicely.</description><pubDate>Thu, 09 Apr 2026 00:00:00 GMT</pubDate><category>AI</category><category>Automation</category><category>n8n</category><category>Content Marketing</category><author>Kyler Berry</author></item><item><title>I Spent $330 on an OpenClaw Content Agent. Here&apos;s What I Got.</title><link>https://kylerberry.com/blog/posts/i-spent-330-on-an-openclaw-content-agent-heres-what-i-got.md</link><guid isPermaLink="true">https://kylerberry.com/blog/posts/i-spent-330-on-an-openclaw-content-agent-heres-what-i-got.md</guid><description>I built a 4-agent LinkedIn content pod on OpenClaw for an honest technical evaluation. Here&apos;s what worked, what broke, and why the cost math doesn&apos;t add up.</description><pubDate>Mon, 23 Mar 2026 00:00:00 GMT</pubDate><category>AI</category><category>Automation</category><category>Content Marketing</category><author>Kyler Berry</author></item><item><title>Why I Scrapped My OpenClaw Content Pod and Rebuilt It in n8n</title><link>https://kylerberry.com/blog/posts/why-i-scrapped-my-openclaw-content-pod-and-rebuilt-it-in-n8n.md</link><guid isPermaLink="true">https://kylerberry.com/blog/posts/why-i-scrapped-my-openclaw-content-pod-and-rebuilt-it-in-n8n.md</guid><description>A spreadsheet audit of my Anthropic API costs revealed the real problem with agent-native automation. Here&apos;s what I rebuilt and the rule for evaluating AI agents.</description><pubDate>Mon, 23 Mar 2026 00:00:00 GMT</pubDate><category>AI</category><category>Automation</category><category>n8n</category><category>Content Marketing</category><author>Kyler Berry</author></item><item><title>Building a Development Documentation MCP Server at Vimeo</title><link>https://kylerberry.com/blog/posts/building-a-development-documentation-mcp-server-at-vimeo.md</link><guid isPermaLink="true">https://kylerberry.com/blog/posts/building-a-development-documentation-mcp-server-at-vimeo.md</guid><description>I built an MCP server that provides AI coding assistants with development documentation for Vimeo&apos;s custom codebase patterns, conventions, and libraries resulting in roughly 75% better output.</description><pubDate>Mon, 29 Sep 2025 00:00:00 GMT</pubDate><category>AI</category><category>MCP</category><category>Case Study</category><author>Kyler Berry</author></item><item><title>One User, One Banana: What My Failed AI Startup Taught Me About Validation</title><link>https://kylerberry.com/blog/posts/one-user-one-banana-what-my-failed-ai-startup-taught-me-about-validation.md</link><guid isPermaLink="true">https://kylerberry.com/blog/posts/one-user-one-banana-what-my-failed-ai-startup-taught-me-about-validation.md</guid><description>I spent 5 months building an AI startup, got 50 signups, and watched my only active user upload a banana. Here&apos;s what it taught me about enthusiasm vs. actual demand.</description><pubDate>Mon, 22 Sep 2025 00:00:00 GMT</pubDate><category>AI</category><category>Case Study</category><author>Kyler Berry</author></item></channel></rss>