<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Serhii Panchyshyn’s Blog</title><description>Fractional CTO and AI engineer for private equity. One AI pilot live in 4 weeks.</description><link>https://serhii.pro</link><item><title>Your own Claude server, off the internet and reachable from anywhere</title><link>https://serhii.pro/blog/own-claude-server</link><guid isPermaLink="true">https://serhii.pro/blog/own-claude-server</guid><description>Rent a computer that runs Claude Code all day, shut every door on it, and steer it from your phone. About $24 a month, every file included.</description><pubDate>Tue, 17 Nov 2026 00:00:00 GMT</pubDate></item><item><title>Build your own apps with Claude, even if you&apos;ve never coded</title><link>https://serhii.pro/blog/build-apps-with-claude</link><guid isPermaLink="true">https://serhii.pro/blog/build-apps-with-claude</guid><description>Four prompts, one folder per idea, and more than one app running by the end of the day. Every prompt is included.</description><pubDate>Wed, 11 Nov 2026 00:00:00 GMT</pubDate></item><item><title>Why Your AI Agent Costs 7x What It Should</title><link>https://serhii.pro/blog/agent-prompt-caching</link><guid isPermaLink="true">https://serhii.pro/blog/agent-prompt-caching</guid><description>Most AI agents re-process the same tokens on every loop iteration. A small layout change to your prompt can cut your input bill by 80%.</description><pubDate>Tue, 21 Apr 2026 00:00:00 GMT</pubDate></item><item><title>No Evals, No Idea. How 40% of RAG Answers Go Wrong.</title><link>https://serhii.pro/blog/rag-evals</link><guid isPermaLink="true">https://serhii.pro/blog/rag-evals</guid><description>Most RAG failures aren&apos;t LLM failures. They&apos;re retrieval failures. Here&apos;s how to measure Recall@k, add hybrid search, and go from 60% to 80% accuracy.</description><pubDate>Sun, 19 Apr 2026 00:00:00 GMT</pubDate></item><item><title>Your Agent Isn&apos;t Broken Because of the Prompt. It&apos;s Broken Because of What the Model Can See.</title><link>https://serhii.pro/blog/engineering-perception</link><guid isPermaLink="true">https://serhii.pro/blog/engineering-perception</guid><description>Teams spend weeks rewriting system prompts. Then they rename a few functions and accuracy jumps 30%. 10 architecture patterns that fix agent quality.</description><pubDate>Fri, 17 Apr 2026 00:00:00 GMT</pubDate></item><item><title>Things You&apos;re Overengineering in Your AI Agent (The LLM Already Handles Them)</title><link>https://serhii.pro/blog/overengineering-agents</link><guid isPermaLink="true">https://serhii.pro/blog/overengineering-agents</guid><description>Teams overengineer routing, memory, guardrails, and error handling. The LLM already handles most of it. Stop building what you don&apos;t need.</description><pubDate>Wed, 15 Apr 2026 00:00:00 GMT</pubDate></item><item><title>The One Mindset Shift That Separates People Who Use AI From People Who Get Left Behind</title><link>https://serhii.pro/blog/curiosity-gap</link><guid isPermaLink="true">https://serhii.pro/blog/curiosity-gap</guid><description>Most people accept friction in their lives without questioning it. The ones thriving in the AI era don&apos;t have better tools. They have better questions.</description><pubDate>Mon, 13 Apr 2026 00:00:00 GMT</pubDate></item></channel></rss>