<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0"><channel>
  <title>Alexander Ryan. Tech Blog</title>
  <link>https://alexryan.tech/</link>
  <description>Essays on AI systems that actually ship to production, by Alexander Ryan, CEO of Ryshe.</description>
  <item>
    <title>Context Engineering Is Not Prompt Engineering with a Better Name</title>
    <link>https://alexryan.tech/blog/context-engineering-discipline.html</link>
    <guid>https://alexryan.tech/blog/context-engineering-discipline.html</guid>
    <pubDate>Wed, 07 Oct 2026 12:00:00 +0000</pubDate>
    <description>Prompt wording is the smallest lever in the system. What the model sees, in what order, under what budget: that is the job now, and it has rules worth learning.</description>
  </item>
  <item>
    <title>Your LLM Evals Are Lying to You</title>
    <link>https://alexryan.tech/blog/llm-evaluation-production-guide.html</link>
    <guid>https://alexryan.tech/blog/llm-evaluation-production-guide.html</guid>
    <pubDate>Tue, 06 Oct 2026 12:00:00 +0000</pubDate>
    <description>Your eval suite passes. Your users still get garbage. The metrics that mean something, the framework landscape without the marketing, and the build order that closes the gap.</description>
  </item>
  <item>
    <title>RAG vs. CAG vs. KAG: Pick the Boring One</title>
    <link>https://alexryan.tech/blog/rag-vs-cag-vs-kag.html</link>
    <guid>https://alexryan.tech/blog/rag-vs-cag-vs-kag.html</guid>
    <pubDate>Mon, 05 Oct 2026 12:00:00 +0000</pubDate>
    <description>Three acronyms, one old question: where should your system&amp;#39;s knowledge live? A production decision framework that ignores the branding.</description>
  </item>
  <item>
    <title>The $400B Lie: Why Most Enterprise AI Projects Fail Before They Start</title>
    <link>https://alexryan.tech/blog/enterprise-ai-projects-fail.html</link>
    <guid>https://alexryan.tech/blog/enterprise-ai-projects-fail.html</guid>
    <pubDate>Sat, 14 Feb 2026 12:00:00 +0000</pubDate>
    <description>After leading AI transformations at multiple enterprises, I've found the same fatal mistakes every time. The problem isn't the technology; it's the question you're asking it to answer.</description>
  </item>
  <item>
    <title>Multi-Agent Systems Are the New Microservices. And Nobody's Ready</title>
    <link>https://alexryan.tech/blog/multi-agent-systems.html</link>
    <guid>https://alexryan.tech/blog/multi-agent-systems.html</guid>
    <pubDate>Sat, 07 Feb 2026 00:00:00 +0000</pubDate>
    <description>Multi-agent AI systems carry every failure mode microservices had, plus new ones. A practitioner's guide to orchestration, observability, and production deployment.</description>
  </item>
  <item>
    <title>I Replaced Our Entire Data Pipeline with an LLM. Here's What Broke.</title>
    <link>https://alexryan.tech/blog/llm-data-pipeline.html</link>
    <guid>https://alexryan.tech/blog/llm-data-pipeline.html</guid>
    <pubDate>Thu, 29 Jan 2026 12:00:00 +0000</pubDate>
    <description>A first-person post-mortem on replacing a traditional ETL pipeline with LLM-powered processing. What worked, what exploded in cost, what hallucinated, and the hybrid architecture we finally settled on.</description>
  </item>
  <item>
    <title>RAG Is Dead. Long Live Agentic Retrieval.</title>
    <link>https://alexryan.tech/blog/rag-is-dead-agentic-retrieval.html</link>
    <guid>https://alexryan.tech/blog/rag-is-dead-agentic-retrieval.html</guid>
    <pubDate>Sun, 18 Jan 2026 00:00:00 +0000</pubDate>
    <description>Vanilla RAG has hit its ceiling. Explore agentic retrieval-agents that plan, reason, and use multi-hop strategies to outperform standard RAG pipelines by 40%+.</description>
  </item>
  <item>
    <title>The AI-Native Frontend: How ML Models Are Eating the UI Layer</title>
    <link>https://alexryan.tech/blog/ai-native-frontend.html</link>
    <guid>https://alexryan.tech/blog/ai-native-frontend.html</guid>
    <pubDate>Tue, 06 Jan 2026 12:00:00 +0000</pubDate>
    <description>ML models are no longer just backends; they're reshaping the UI layer itself. Predictive prefetching, adaptive layouts, and AI-generated components are here.</description>
  </item>
  <item>
    <title>16 Years in Tech: The Only 5 Things That Actually Mattered</title>
    <link>https://alexryan.tech/blog/16-years-in-tech.html</link>
    <guid>https://alexryan.tech/blog/16-years-in-tech.html</guid>
    <pubDate>Mon, 22 Dec 2025 12:00:00 +0000</pubDate>
    <description>From junior dev in Detroit to CEO of Ryshe; 16 years of hard-won lessons distilled into the only 5 things that actually changed the trajectory of my career.</description>
  </item>
  <item>
    <title>Fine-Tuning is a Trap. Here's What to Do Instead.</title>
    <link>https://alexryan.tech/blog/fine-tuning-is-a-trap.html</link>
    <guid>https://alexryan.tech/blog/fine-tuning-is-a-trap.html</guid>
    <pubDate>Wed, 10 Dec 2025 12:00:00 +0000</pubDate>
    <description>After 8 failed fine-tuning experiments and $40K in GPU costs, here's what actually works: prompt engineering, RAG, few-shot learning, and a decision framework for when fine-tuning is genuinely the right call.</description>
  </item>
</channel></rss>
