<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Ithaca</title><description>Technical notes on AI engineering, LLM systems, agents, and evaluation.</description><link>https://qianxinchun.com/</link><item><title>Stop Prompting Coding Agents. Start Designing Loops.</title><link>https://qianxinchun.com/blog/stop-prompting-coding-agents-start-designing-loops/</link><guid isPermaLink="true">https://qianxinchun.com/blog/stop-prompting-coding-agents-start-designing-loops/</guid><description>The real shift in AI coding is not from bad prompts to better prompts. It is from human-driven conversations to systems that can observe, verify, remember, and decide when to ask the agent again.</description><pubDate>Tue, 09 Jun 2026 00:00:00 GMT</pubDate><category>Agents</category><category>LLMs</category><category>AI Engineering</category><category>Coding Agents</category><category>Loop Engineering</category></item><item><title>Why I&apos;m Writing About LLM Systems and Agents</title><link>https://qianxinchun.com/blog/why-im-writing-about-llm-systems-and-agents/</link><guid isPermaLink="true">https://qianxinchun.com/blog/why-im-writing-about-llm-systems-and-agents/</guid><description>A first note on building a public learning practice around LLM systems, agents, evaluation, and the engineering habits behind them.</description><pubDate>Tue, 09 Jun 2026 00:00:00 GMT</pubDate><category>LLMs</category><category>Agents</category><category>AI Engineering</category><category>Writing</category></item></channel></rss>