<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Conversational-Ai on simalexan</title><link>https://simalexan.com/tags/conversational-ai/</link><description>Recent content in Conversational-Ai on simalexan</description><generator>Hugo -- gohugo.io</generator><language>en-us</language><lastBuildDate>Thu, 22 Oct 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://simalexan.com/tags/conversational-ai/index.xml" rel="self" type="application/rss+xml"/><item><title>Journey to Intelligent Conversational Agents</title><link>https://simalexan.com/presentations/journey-to-intelligent-conversational-agents/</link><pubDate>Thu, 22 Oct 2026 00:00:00 +0000</pubDate><guid>https://simalexan.com/presentations/journey-to-intelligent-conversational-agents/</guid><description>Why AI has us working more rather than less, what a multimodal voice agent is made of, and how we built one: Ultron.</description><content:encoded>&lt;p>A talk Konstantinos Siaterlis and I gave at AWS Community Day Adria 2026 in Ljubljana, Slovenia, on October 22, 2026: why AI has us working more rather than less, what a voice agent is made of, and how we built Ultron, a multimodal voice agent that listens to a design discussion and draws on a board.&lt;/p>
&lt;figure class="my-10">
&lt;iframe
src="https://simalexan.com/presentations/journey-to-intelligent-conversational-agents/deck/"
title="Journey to Intelligent Conversational Agents, the slides"
allow="fullscreen"
allowfullscreen
style="display: block; width: 100%; aspect-ratio: 16 / 9; border: 0; border-radius: 0.5rem; background: #01010a"
>&lt;/iframe>
&lt;figcaption class="mt-3 text-sm">
Click the slides, then use ← and → to move and F for full screen.
&lt;a href="https://simalexan.com/presentations/journey-to-intelligent-conversational-agents/deck/">Open the slides on their own page →&lt;/a>
&lt;/figcaption>
&lt;/figure>
&lt;h2 id="what-it-covers">What it covers&lt;/h2>
&lt;ul>
&lt;li>&lt;strong>Opening&lt;/strong>: why we now work more, not less, with AI (the Jevons paradox, and &amp;ldquo;botsitting&amp;rdquo; ten agents), and whether voice agents are a cure.&lt;/li>
&lt;li>&lt;strong>Strands&lt;/strong>: building voice agents with Strands and Amazon Nova Sonic, from ten lines of code to an agent that researches a topic and emails it to you.&lt;/li>
&lt;li>&lt;strong>Origins&lt;/strong>: from Turing to ChatGPT, by way of the re:Invent 2018 workshop &amp;ldquo;Alexa, ask Jarvis to create a serverless app for me&amp;rdquo;.&lt;/li>
&lt;li>&lt;strong>Anatomy&lt;/strong>: what a multimodal voice agent is made of, and why we chose each part of Ultron.&lt;/li>
&lt;li>&lt;strong>Close&lt;/strong>: Ultron drawing its own architecture, Wardley Maps, and three key points.&lt;/li>
&lt;/ul>
&lt;h2 id="three-key-points">Three key points&lt;/h2>
&lt;ul>
&lt;li>&lt;strong>Voice in, pictures out&lt;/strong>: speak to tell it, see to understand it, both ways in one conversation.&lt;/li>
&lt;li>&lt;strong>Built from parts&lt;/strong>: diarization, interruptions, speech, a voice and memory, each chosen on evidence.&lt;/li>
&lt;li>&lt;strong>Fewer calls, more decisions&lt;/strong>: an agent that calls you only to decide cuts the waiting and the checking.&lt;/li>
&lt;/ul></content:encoded><category>conversational-ai</category><category>voice-agents</category><category>ai-agents</category><category>strands</category><category>wardley-mapping</category></item></channel></rss>