https://carlos.enredando.me/tags/reasoning/Recent content in Reasoning on EnRedAndo Me - Carlos PradosHugo -- gohugo.ioenmail@carlosprados.com (Carlos Prados)mail@carlosprados.com (Carlos Prados)© 2026 Carlos PradosTue, 01 Sep 2026 09:00:00 +0200https://carlos.enredando.me/posts/agentic-ai-reasoning/Tue, 01 Sep 2026 09:00:00 +0200mail@carlosprados.com (Carlos Prados)https://carlos.enredando.me/posts/agentic-ai-reasoning/<p>In my <a href="https://carlos.enredando.me/posts/agentic-ai-resource-aware/" >previous post</a>, we looked at Resource-Aware Optimization — teaching agents to spend tokens, latency, and money deliberately instead of burning the biggest model on every request. That pattern is about <em>how much</em> to think. This one is about <em>how</em> to think.</p> <p>Most LLM failures on hard problems aren&rsquo;t knowledge failures. The model knows the facts. It just blurts out the answer in one shot, skips the intermediate steps, and gets the logic wrong. Ask it to compute compound interest or untangle a multi-hop question and it will confidently hand you a plausible-looking mistake.</p>