<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Cabinly Blog</title><description>Plain notes on AI agents, LLM products and how a small team ships software.</description><link>https://cabinlytech.com/</link><language>en</language><item><title>How we ship software with AI agents without losing control</title><link>https://cabinlytech.com/blog/shipping-software-with-ai-agents/</link><guid isPermaLink="true">https://cabinlytech.com/blog/shipping-software-with-ai-agents/</guid><description>Specs, small plans, task-scoped agents, independent review and real checks. How agents do the work while people keep the decisions.</description><pubDate>Tue, 29 Sep 2026 00:00:00 GMT</pubDate><category>agents</category><category>engineering</category><category>process</category></item><item><title>Routing between LLM providers to balance cost and quality</title><link>https://cabinlytech.com/blog/routing-between-llm-providers/</link><guid isPermaLink="true">https://cabinlytech.com/blog/routing-between-llm-providers/</guid><description>Why we route requests across more than one model provider, how prompt caching cuts repeat costs, and why routing changes pass offline evals first.</description><pubDate>Tue, 22 Sep 2026 00:00:00 GMT</pubDate><category>llm</category><category>cost</category><category>evals</category></item><item><title>Designing AI memory that stays accurate</title><link>https://cabinlytech.com/blog/designing-ai-memory-that-stays-accurate/</link><guid isPermaLink="true">https://cabinlytech.com/blog/designing-ai-memory-that-stays-accurate/</guid><description>Long-term memory for an AI companion is about keeping facts current, not storing more. Recall, supersession, and why contradictions cost trust.</description><pubDate>Tue, 15 Sep 2026 00:00:00 GMT</pubDate><category>llm</category><category>memory</category><category>product</category></item></channel></rss>