<?xml version="1.0" encoding="UTF-8" ?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Halios Blog — Engineering &amp; Research</title>
    <link>https://halios.ai/blog</link>
    <description>Technical notes on AI agent evaluations, multi-turn simulations, OpenTelemetry, and building resilient systems.</description>
    <language>en-us</language>
    <lastBuildDate>Thu, 17 Sep 2026 10:27:02 GMT</lastBuildDate>
    <atom:link href="https://halios.ai/feed.xml" rel="self" type="application/rss+xml" />
    
    <item>
      <title><![CDATA[The Real Cost of Building Agent Eval Infrastructure]]></title>
      <link>https://halios.ai/blog/the-real-cost-of-building-agent-eval-infrastructure</link>
      <guid isPermaLink="true">https://halios.ai/blog/the-real-cost-of-building-agent-eval-infrastructure</guid>
      <description><![CDATA[Writing a basic eval is easy. The work grows when you want to evaluate a multi-turn agent repeatedly and reliably across environments.]]></description>
      <pubDate>Sun, 23 Aug 2026 00:00:00 GMT</pubDate>
      <author>Sandeep B</author>
      <category>Agent Evals</category>
      <category>Architecture</category>
      <category>Build vs Buy</category>
      <category>Infrastructure</category>
    </item>
    <item>
      <title><![CDATA[How Coding Agents Build Better AI Agent Evaluations]]></title>
      <link>https://halios.ai/blog/coding-agents-build-ai-agent-evaluations</link>
      <guid isPermaLink="true">https://halios.ai/blog/coding-agents-build-ai-agent-evaluations</guid>
      <description><![CDATA[A coding agent already understands your tools, prompts, policies, and application structure. Halios turns that context into a practical workflow for creating and maintaining agent evals.]]></description>
      <pubDate>Sat, 22 Aug 2026 00:00:00 GMT</pubDate>
      <author>Sandeep B</author>
      <category>Coding Agents</category>
      <category>Agent Evals</category>
      <category>Workflow</category>
    </item>
    <item>
      <title><![CDATA[AI Agent Evals from Local Development to CI and Production]]></title>
      <link>https://halios.ai/blog/ai-agent-evals-local-ci-production</link>
      <guid isPermaLink="true">https://halios.ai/blog/ai-agent-evals-local-ci-production</guid>
      <description><![CDATA[Use one evaluation model across local iteration, pull-request quality gates, and production behavior instead of maintaining three disconnected testing systems.]]></description>
      <pubDate>Fri, 21 Aug 2026 00:00:00 GMT</pubDate>
      <author>Halios Team</author>
      <category>CI/CD</category>
      <category>Agent Evals</category>
      <category>Production</category>
    </item>
    <item>
      <title><![CDATA[Why Halios Uses Scenarios Instead of Eval Datasets]]></title>
      <link>https://halios.ai/blog/why-halios-uses-scenarios-instead-of-eval-datasets</link>
      <guid isPermaLink="true">https://halios.ai/blog/why-halios-uses-scenarios-instead-of-eval-datasets</guid>
      <description><![CDATA[Most evaluation tools start with a dataset. Why Halios makes scenarios the primary unit of evaluation for multi-turn AI agents.]]></description>
      <pubDate>Thu, 20 Aug 2026 00:00:00 GMT</pubDate>
      <author>Halios Team</author>
      <category>Scenarios</category>
      <category>Simulations</category>
      <category>LLM Evaluation</category>
    </item>
    <item>
      <title><![CDATA[Why AI Agent Evals Belong in Your Git Repository]]></title>
      <link>https://halios.ai/blog/why-ai-agent-evals-belong-in-git</link>
      <guid isPermaLink="true">https://halios.ai/blog/why-ai-agent-evals-belong-in-git</guid>
      <description><![CDATA[Keeping test scenarios and checks alongside your agent's code gives coding agents context, streamlines PR reviews, and avoids dashboard silos.]]></description>
      <pubDate>Wed, 19 Aug 2026 00:00:00 GMT</pubDate>
      <author>Halios Team</author>
      <category>Git</category>
      <category>Eval as Code</category>
      <category>Developer Workflow</category>
    </item>
    <item>
      <title><![CDATA[OpenTelemetry for Portable AI Agent Evaluation Traces]]></title>
      <link>https://halios.ai/blog/opentelemetry-ai-agent-evaluation-traces</link>
      <guid isPermaLink="true">https://halios.ai/blog/opentelemetry-ai-agent-evaluation-traces</guid>
      <description><![CDATA[Why proprietary eval SDKs are technical debt, and how standard OpenTelemetry GenAI traces give you portable agent evaluations without lock-in.]]></description>
      <pubDate>Tue, 18 Aug 2026 00:00:00 GMT</pubDate>
      <author>Halios Team</author>
      <category>OpenTelemetry</category>
      <category>Agent Tracing</category>
      <category>Portability</category>
    </item>
    <item>
      <title><![CDATA[How to Build a Continual AI Agent Improvement Loop]]></title>
      <link>https://halios.ai/blog/continual-ai-agent-evaluation-improvement-loop</link>
      <guid isPermaLink="true">https://halios.ai/blog/continual-ai-agent-evaluation-improvement-loop</guid>
      <description><![CDATA[How to turn agent evaluations from a passive dashboard metric into an active coding loop with your coding agent and CI.]]></description>
      <pubDate>Mon, 17 Aug 2026 00:00:00 GMT</pubDate>
      <author>Sandeep B</author>
      <category>Continual Improvement</category>
      <category>Agent Reliability</category>
      <category>Regression Testing</category>
    </item>
  </channel>
</rss>