https://peter.truchly.eu/tags/systemdesign/Recent content in Systemdesign on Peter TruchlyHugo -- gohugo.ioenMon, 17 Aug 2026 00:00:00 +0200https://peter.truchly.eu/p/why-llm-chat-feels-like-an-anti-pattern/Fri, 24 Jul 2026 00:00:00 +0200https://peter.truchly.eu/p/why-llm-chat-feels-like-an-anti-pattern/<img src="https://peter.truchly.eu/p/why-llm-chat-feels-like-an-anti-pattern/banner-b11_hu_ad8fcce89afb12da.png" alt="Featured image of post Why LLM chat feels like an anti-pattern in 2026" /><p><em>A continuation of my previous article, <a class="link" href="https://peter.truchly.eu/p/what-is-better-than-your-ai-loop/" >What is better than your AI loop?</a>, which introduced the blackboard architecture this post builds on.</em></p> <p>The <strong>chat</strong>. The message-response chain you produce by typing with an LLM. It could be trivially implemented by a single growing context or a sliding window.</p> <p>Why does it feel like an anti-pattern to me? Chat as a UI is fine, chat as a world model is not.</p>https://peter.truchly.eu/p/what-is-better-than-your-ai-loop/Mon, 06 Jul 2026 00:00:00 +0200https://peter.truchly.eu/p/what-is-better-than-your-ai-loop/<img src="https://peter.truchly.eu/p/what-is-better-than-your-ai-loop/cover_hu_3c3546198588fe5c.png" alt="Featured image of post What is better than your AI loop?" /><p>There is a continuation of this article, <a class="link" href="https://peter.truchly.eu/p/why-llm-chat-feels-like-an-anti-pattern/" >Why LLM chat feels like an anti-pattern in 2026?</a>, which shows the blackboard architecture in action.</p> <p>A blackboard architecture, modernized with event sourcing, content-addressed knowledge source (KS) firing and a few more enhancements.</p> <p>While working on improvements to various LLM-powered components and subsystems over the past year, I noticed I kept converging on the same architecture — one that, as I later found out, had been described and named four decades ago.</p>