<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:itunes="http://www.itunes.com/dtds/podcast-1.0.dtd" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:sy="http://purl.org/rss/1.0/modules/syndication/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:podcast="https://podcastindex.org/namespace/1.0" version="2.0">
  <channel>
    <atom:link rel="self" type="application/rss+xml" href="https://feeds.beamly.com/GBGPGHcv3Ozii49J/feed.xml" title="MP3 Audio"/>
    <generator>Beamly.com</generator>
    <title>Agent Driven Development</title>
    <description>A podcast about coding with agents, building agents, and trying to keep the humans in the loop of what's going on in the world of AI.</description>
    <itunes:summary>A podcast about coding with agents, building agents, and trying to keep the humans in the loop of what's going on in the world of AI.</itunes:summary>
    <link>https://agentdrivendevelopment.fm</link>
    <lastBuildDate>Sat, 02 May 2026 01:45:51 GMT</lastBuildDate>
    <podcast:guid>42d24d7e-b643-5c5f-a055-0dbbb0abf552</podcast:guid>
    <language>en</language>
    <itunes:explicit>false</itunes:explicit>
    <itunes:complete>no</itunes:complete>
    <copyright>© 2026. Scott Sauber</copyright>
    <itunes:author>Scott Sauber</itunes:author>
    <itunes:subtitle>A podcast about coding with agents, building agents, and trying to keep the humans in the loop of what's going on in the world of AI.</itunes:subtitle>
    <itunes:keywords>Artificial Intelligence, AI, Agentic Coding, Agents, Agent Driven Development, Programming, Software Development, Software Engineering</itunes:keywords>
    <itunes:owner>
      <itunes:name>Scott Sauber</itunes:name>
    </itunes:owner>
    <image>
      <url>https://sites.beamly.com/69dbc8ec31a1ae88eac8647a/media/38d2e1e97869531d9ac6.jpg</url>
      <title>Agent Driven Development</title>
      <link>https://agentdrivendevelopment.fm</link>
    </image>
    <itunes:image href="https://sites.beamly.com/69dbc8ec31a1ae88eac8647a/media/38d2e1e97869531d9ac6.jpg"/>
    <itunes:category text="Technology"/>
    <item>
      <title>Microsoft Agent Framework with Shawn Henry</title>
      <itunes:title>Microsoft Agent Framework with Shawn Henry</itunes:title>
      <pubDate>Thu, 30 Apr 2026 01:06:00 GMT</pubDate>
      <itunes:episode>001</itunes:episode>
      <podcast:episode>001</podcast:episode>
      <guid isPermaLink="false">ddf9fe7e-f8ce-4888-a7f3-effd8a3cdd60</guid>
      <link>https://agentdrivendevelopment.fm/podcast/agent-driven-development/episode/microsoft-agent-framework-with-shawn-henry</link>
      <description><![CDATA[<p><span style="color: #05e5bf">Microsoft Agent Framework is designed to help teams build reliable AI and agent applications with the same engineering discipline they already use in .NET and cloud development. In this conversation, hosts Scott Sauber and Spencer Schneidenbach talk with Shawn Henry, Program Manager for Microsoft Agent Framework, about where the framework came from, why it exists, and how it helps developers move from experiments to production-ready agent systems.</span></p><p><span style="color: #05e5bf">Shawn shares his path through Microsoft—from Windows Phone and Azure Communication Services to Semantic Kernel, AutoGen, and now Microsoft Agent Framework—and explains how those efforts came together to create a more unified approach for building agents. The discussion covers migration from Semantic Kernel, the role of tools, middleware, compaction, memory, and how the framework helps developers avoid the common traps of overly large prompts, tool overload, and brittle orchestration logic.</span></p><p><span style="color: #05e5bf">If you’re working in .NET, Python, or enterprise AI more broadly, this episode offers a practical look at how Microsoft is thinking about agentic development, scale, and the developer experience.</span></p><h3 id="5ea2e8" class=""><span style="color: #ffffff">Key Topics</span></h3><p><span style="color: #6a5ef0">[00:00:00]</span><span style="color: #05e5bf"> - </span><span style="color: #ffffff">Introductions with Scott Sauber, Spencer Schneidenbach, and Shawn Henry</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:00:19]</span><span style="color: #05e5bf"> - </span><span style="color: #ffffff">Shawn Henry’s Microsoft journey across Windows Phone, Azure, Semantic Kernel, and AI</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:02:29] </span><span style="color: #05e5bf">- </span><span style="color: #ffffff">What Microsoft Agent Framework is and the problems it solves</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:04:30] </span><span style="color: #05e5bf">- </span><span style="color: #ffffff">Agents, prompts, models, and tools as the core building blocks</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:06:18]</span><span style="color: #05e5bf"> - </span><span style="color: #ffffff">Migration path from Semantic Kernel and AutoGen to Agent Framework</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:09:12]</span><span style="color: #05e5bf"> - </span><span style="color: #ffffff">Why Microsoft shifted toward a more agent-centric abstraction</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:11:46] </span><span style="color: #05e5bf">- </span><span style="color: #ffffff">Middleware and structured data as underused but powerful capabilities</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:16:32] </span><span style="color: #05e5bf">- </span><span style="color: #ffffff">Avoiding too many tools, context overload, and weak agent design</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:20:29] </span><span style="color: #05e5bf">- </span><span style="color: #ffffff">Real-world scenarios where agents can add value in existing workflows</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:24:41] </span><span style="color: #05e5bf">- </span><span style="color: #ffffff">Dogfooding: how Microsoft teams use agent framework internally</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:28:37]</span><span style="color: #05e5bf"> - </span><span style="color: #ffffff">VS Code, AI Toolkit, Copilot, and deployment tooling</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:31:00]</span><span style="color: #05e5bf"> - </span><span style="color: #ffffff">What’s next: GA, enterprise scale, and Claude-like systems</span></p><h3 id="8c724" class=""><span style="color: #ffffff">Relevant Links</span></h3><ul class="tight" data-tight="true"><li><p><a target="_blank" class="link" href="https://github.com/microsoft/agent-framework" data-type="link">https://github.com/microsoft/agent-framework</a></p></li><li><p><a target="_blank" class="link" href="https://github.com/microsoft/semantic-kernel" data-type="link">https://github.com/microsoft/semantic-kernel</a></p></li><li><p><a target="_blank" class="link" href="https://github.com/microsoft/autogen" data-type="link">https://github.com/microsoft/autogen</a></p></li><li><p><a target="_blank" class="link" href="" data-type="link">https://code.visualstudio.com/docs/copilot/ai-toolkit</a></p></li><li><p><a target="_blank" class="link" href="https://learn.microsoft.com/en-us/azure/ai-foundry/" data-type="link">https://learn.microsoft.com/en-us/azure/ai-foundry/</a></p></li><li><p><a target="_blank" class="link" href="https://aka.ms/agent-framework" data-type="link">https://aka.ms/agent-framework</a></p></li></ul><p></p><p><span style="color: #05e5bf">Shawn's core message is simple: if you’re already building applications, agent framework helps you extend that code into AI-powered systems without starting over. The conversation also makes a strong case for treating agents as production software—something that benefits from middleware, observability, compaction, and thoughtful tool design.</span></p><p><span style="color: #05e5bf">For teams trying to modernize workflows, reduce toil, or bring AI into existing systems, this episode is a practical starting point.</span></p>]]></description>
      <content:encoded><![CDATA[<p><span style="color: #05e5bf">Microsoft Agent Framework is designed to help teams build reliable AI and agent applications with the same engineering discipline they already use in .NET and cloud development. In this conversation, hosts Scott Sauber and Spencer Schneidenbach talk with Shawn Henry, Program Manager for Microsoft Agent Framework, about where the framework came from, why it exists, and how it helps developers move from experiments to production-ready agent systems.</span></p><p><span style="color: #05e5bf">Shawn shares his path through Microsoft—from Windows Phone and Azure Communication Services to Semantic Kernel, AutoGen, and now Microsoft Agent Framework—and explains how those efforts came together to create a more unified approach for building agents. The discussion covers migration from Semantic Kernel, the role of tools, middleware, compaction, memory, and how the framework helps developers avoid the common traps of overly large prompts, tool overload, and brittle orchestration logic.</span></p><p><span style="color: #05e5bf">If you’re working in .NET, Python, or enterprise AI more broadly, this episode offers a practical look at how Microsoft is thinking about agentic development, scale, and the developer experience.</span></p><h3 id="5ea2e8" class=""><span style="color: #ffffff">Key Topics</span></h3><p><span style="color: #6a5ef0">[00:00:00]</span><span style="color: #05e5bf"> - </span><span style="color: #ffffff">Introductions with Scott Sauber, Spencer Schneidenbach, and Shawn Henry</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:00:19]</span><span style="color: #05e5bf"> - </span><span style="color: #ffffff">Shawn Henry’s Microsoft journey across Windows Phone, Azure, Semantic Kernel, and AI</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:02:29] </span><span style="color: #05e5bf">- </span><span style="color: #ffffff">What Microsoft Agent Framework is and the problems it solves</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:04:30] </span><span style="color: #05e5bf">- </span><span style="color: #ffffff">Agents, prompts, models, and tools as the core building blocks</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:06:18]</span><span style="color: #05e5bf"> - </span><span style="color: #ffffff">Migration path from Semantic Kernel and AutoGen to Agent Framework</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:09:12]</span><span style="color: #05e5bf"> - </span><span style="color: #ffffff">Why Microsoft shifted toward a more agent-centric abstraction</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:11:46] </span><span style="color: #05e5bf">- </span><span style="color: #ffffff">Middleware and structured data as underused but powerful capabilities</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:16:32] </span><span style="color: #05e5bf">- </span><span style="color: #ffffff">Avoiding too many tools, context overload, and weak agent design</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:20:29] </span><span style="color: #05e5bf">- </span><span style="color: #ffffff">Real-world scenarios where agents can add value in existing workflows</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:24:41] </span><span style="color: #05e5bf">- </span><span style="color: #ffffff">Dogfooding: how Microsoft teams use agent framework internally</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:28:37]</span><span style="color: #05e5bf"> - </span><span style="color: #ffffff">VS Code, AI Toolkit, Copilot, and deployment tooling</span><span style="color: #05e5bf"><br></span><span style="color: #6a5ef0">[00:31:00]</span><span style="color: #05e5bf"> - </span><span style="color: #ffffff">What’s next: GA, enterprise scale, and Claude-like systems</span></p><h3 id="8c724" class=""><span style="color: #ffffff">Relevant Links</span></h3><ul class="tight" data-tight="true"><li><p><a target="_blank" class="link" href="https://github.com/microsoft/agent-framework" data-type="link">https://github.com/microsoft/agent-framework</a></p></li><li><p><a target="_blank" class="link" href="https://github.com/microsoft/semantic-kernel" data-type="link">https://github.com/microsoft/semantic-kernel</a></p></li><li><p><a target="_blank" class="link" href="https://github.com/microsoft/autogen" data-type="link">https://github.com/microsoft/autogen</a></p></li><li><p><a target="_blank" class="link" href="" data-type="link">https://code.visualstudio.com/docs/copilot/ai-toolkit</a></p></li><li><p><a target="_blank" class="link" href="https://learn.microsoft.com/en-us/azure/ai-foundry/" data-type="link">https://learn.microsoft.com/en-us/azure/ai-foundry/</a></p></li><li><p><a target="_blank" class="link" href="https://aka.ms/agent-framework" data-type="link">https://aka.ms/agent-framework</a></p></li></ul><p></p><p><span style="color: #05e5bf">Shawn's core message is simple: if you’re already building applications, agent framework helps you extend that code into AI-powered systems without starting over. The conversation also makes a strong case for treating agents as production software—something that benefits from middleware, observability, compaction, and thoughtful tool design.</span></p><p><span style="color: #05e5bf">For teams trying to modernize workflows, reduce toil, or bring AI into existing systems, this episode is a practical starting point.</span></p>]]></content:encoded>
      <itunes:episodeType>full</itunes:episodeType>
      <enclosure url="https://media.beamly.com/GBGM9duNQLIxChS2/GBGPGHcv3Ozii49J/GE35EKnDLRMtIGAm/6712f90408fc.wav" length="292603418" type="audio/wave"/>
      <itunes:image href="https://sites.beamly.com/69dbc8ec31a1ae88eac8647a/media/29a7c1cf5bec5bd5687e.webp"/>
      <itunes:keywords>ai,microsoft,agents,artificial intelligence,microsoft agent framework,agent driven development</itunes:keywords>
      <podcast:transcript url="https://feeds.beamly.com/GBGPGHcv3Ozii49J/transcripts/69f3fe7937df8733f3578ca6.html" type="text/html"/>
      <podcast:transcript url="https://feeds.beamly.com/GBGPGHcv3Ozii49J/transcripts/69f3fe7937df8733f3578ca6.vtt" type="text/vtt"/>
      <itunes:duration>00:33:52</itunes:duration>
      <itunes:summary>In this episode, Scott Sauber and Spencer Schneidenbach talk with Shawn Henry, Program Manager  at Microsoft for Microsoft Agent Framework, about how the framework helps teams build production ready AI agents and agentic applications. They explore the origins of the framework, its migration path from Semantic Kernel and AutoGen, and how middleware, structured data, and careful tool design help avoid common pitfalls.</itunes:summary>
      <itunes:explicit>false</itunes:explicit>
    </item>
  </channel>
</rss>