<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>HOAi</title><description>HOAi Engineering. We build the AI workforce for HOA management. We are hiring.</description><link>https://engineering.myhoai.com/</link><item><title>Debugging stuck Node.js processes</title><link>https://engineering.myhoai.com/posts/debugging-stuck-node-js-processes/</link><guid isPermaLink="true">https://engineering.myhoai.com/posts/debugging-stuck-node-js-processes/</guid><description>One of our critical apps kept failing health checks. Logs, metrics, and local repros came up empty. We dove into Node.js internals to find the root cause.</description><pubDate>Wed, 29 Jul 2026 00:00:00 GMT</pubDate></item><item><title>When the model overthinks</title><link>https://engineering.myhoai.com/posts/when-the-model-overthinks/</link><guid isPermaLink="true">https://engineering.myhoai.com/posts/when-the-model-overthinks/</guid><description>Sometimes reasoning models get stuck in loops that burn thousands of tokens and produce nothing. Here is how we fixed it.</description><pubDate>Tue, 07 Jul 2026 00:00:00 GMT</pubDate></item><item><title>A simple fix for LLM tail latency</title><link>https://engineering.myhoai.com/posts/a-simple-fix-for-llm-tail-latency/</link><guid isPermaLink="true">https://engineering.myhoai.com/posts/a-simple-fix-for-llm-tail-latency/</guid><description>Send every request twice and take the faster response.</description><pubDate>Wed, 10 Jun 2026 00:00:00 GMT</pubDate></item><item><title>Just ask the model why</title><link>https://engineering.myhoai.com/posts/just-ask-the-model-why/</link><guid isPermaLink="true">https://engineering.myhoai.com/posts/just-ask-the-model-why/</guid><description>When your agent does the wrong thing, ask it why. It usually knows.</description><pubDate>Fri, 15 May 2026 00:00:00 GMT</pubDate></item><item><title>How we evaluate AI agents across 1,000+ production scenarios</title><link>https://engineering.myhoai.com/posts/how-we-evaluate-ai-agents/</link><guid isPermaLink="true">https://engineering.myhoai.com/posts/how-we-evaluate-ai-agents/</guid><description>Generic benchmarks can&apos;t tell you if your AI agent will make the right decisions in production. We built a replay-based evaluation framework that tests 1,000+ real scenarios to compare models and catch regressions before they reach 200,000 weekly workflow runs.</description><pubDate>Tue, 07 Apr 2026 00:00:00 GMT</pubDate></item><item><title>How I run parallel coding agents</title><link>https://engineering.myhoai.com/posts/how-i-run-parallel-coding-agents/</link><guid isPermaLink="true">https://engineering.myhoai.com/posts/how-i-run-parallel-coding-agents/</guid><description>Lessons from running multiple Claude Code agents at HOAi</description><pubDate>Fri, 03 Apr 2026 00:00:00 GMT</pubDate></item></channel></rss>