<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:itunes="http://www.itunes.com/dtds/podcast-1.0.dtd" xmlns:psc="http://podlove.org/simple-chapters" xmlns:podcast="https://podcastindex.org/namespace/1.0"><channel><title><![CDATA[Agent Hour]]></title><description><![CDATA[<p>Agent Hour is a show for people building in the AI agent space — engineers, founders, and technical leaders who are actually shipping this stuff. The pod goes a mix of wide and deep to share knowledge related to the industry as it continues to be revealed. And true to the premise of agents improving our lives, every episode runs under 60 minutes, because if the show can't practice what it preaches about agents saving time, what's the point. </p><p></p><p>Agent Hour is sponsored by <a rel="noopener noreferrer nofollow" href="www.anchorbrowser.io" target="_blank">Anchor Browser </a>the enterprise-grade browser purpose-built for agents to surf the web and take action. </p>]]></description><link>https://podcasters.spotify.com/pod/show/gabi-m-weinberg7</link><generator>Riverside.fm (https://riverside.com)</generator><lastBuildDate>Fri, 11 Sep 2026 05:48:06 GMT</lastBuildDate><atom:link href="https://api.riverside.com/hosting/zX35fgXv.rss" rel="self" type="application/rss+xml"/><author><![CDATA[Anchor Browser]]></author><pubDate>Wed, 29 Jul 2026 17:15:05 GMT</pubDate><copyright><![CDATA[2026 Anchor Browser]]></copyright><language><![CDATA[en]]></language><ttl>60</ttl><category><![CDATA[Technology]]></category><itunes:author>Anchor Browser</itunes:author><itunes:summary>&lt;p&gt;Agent Hour is a show for people building in the AI agent space — engineers, founders, and technical leaders who are actually shipping this stuff. The pod goes a mix of wide and deep to share knowledge related to the industry as it continues to be revealed. And true to the premise of agents improving our lives, every episode runs under 60 minutes, because if the show can&apos;t practice what it preaches about agents saving time, what&apos;s the point. &lt;/p&gt;&lt;p&gt;&lt;/p&gt;&lt;p&gt;Agent Hour is sponsored by &lt;a rel=&quot;noopener noreferrer nofollow&quot; href=&quot;www.anchorbrowser.io&quot; target=&quot;_blank&quot;&gt;Anchor Browser &lt;/a&gt;the enterprise-grade browser purpose-built for agents to surf the web and take action. &lt;/p&gt;</itunes:summary><itunes:type>episodic</itunes:type><itunes:owner><itunes:name>Anchor Browser</itunes:name><itunes:email>gabi@anchorbrowser.io</itunes:email></itunes:owner><itunes:explicit>no</itunes:explicit><itunes:category text="Technology"/><itunes:image href="https://hosting-media.riverside.com/media/imports/podcasts/869e72ce-82ce-46e5-9aef-a068c3d4a4b7/46436792-1785160819428-6082b6d2fd3cb.jpg"/><item><title><![CDATA[Episode 004 - Vasilije Markovic, CEO // Cognee]]></title><description><![CDATA[<p>When an agent's session ends, its context usually goes with it — and that problem multiplies fast once you've got multiple agents running at once, each with different access to different data. Vas, founder and CEO of Cognee, joins Agent Hour to explain why simple vector search isn't enough for real agent memory, and how knowledge graphs let agents actually reason over relationships instead of just retrieving similar chunks. He walks through Cognee's ECL pipeline (Extract, Cognify, Load) for turning messy multi-source data into something agents can navigate, plus how granular, per-agent access control keeps each agent's "world model" clean as teams scale.</p><p></p><p>🎧 Brought to you by <a rel="noopener noreferrer nofollow" href="https://anchorbrowser.io" target="_blank">Anchor Browser</a> — cloud-native browser infrastructure for AI agents.</p>]]></description><guid isPermaLink="false">986ac35a-26fe-4fff-84b0-bc271d198b20</guid><dc:creator><![CDATA[Anchor Browser]]></dc:creator><pubDate>Tue, 01 Sep 2026 20:32:42 GMT</pubDate><enclosure url="https://api.riverside.com/hosting-analytics/media/2b093c7c900cbcbb37487b66d5a6bd5000fd477c8c9559022dcda15565b3ce8f/eyJlcGlzb2RlSWQiOiI5ODZhYzM1YS0yNmZlLTRmZmYtODRiMC1iYzI3MWQxOThiMjAiLCJwb2RjYXN0SWQiOiI4NjllNzJjZS04MmNlLTQ2ZTUtOWFlZi1hMDY4YzNkNGE0YjciLCJhY2NvdW50SWQiOiI2OWZjOGNhY2M1MzU3NDUxMTE4NmU5ODciLCJwYXRoIjoibWVkaWEvY2xpcHMvNmE4MzMyODJlMTg3YmUwZTc5ZjRkNTM5L2dhYmlzLXN0dWRpby15a2doWS1jb21wb3Nlci0yMDI2LTgtMTdfXzE4LTEwLTQyLm1wMyJ9.mp3" length="46615135" type="audio/mpeg"/><podcast:transcript url="https://hosting-media.riverside.com/media/podcasts/869e72ce-82ce-46e5-9aef-a068c3d4a4b7/episodes/986ac35a-26fe-4fff-84b0-bc271d198b20/transcripts.txt" type="text/plain"/><itunes:summary>&lt;p&gt;When an agent&apos;s session ends, its context usually goes with it — and that problem multiplies fast once you&apos;ve got multiple agents running at once, each with different access to different data. Vas, founder and CEO of Cognee, joins Agent Hour to explain why simple vector search isn&apos;t enough for real agent memory, and how knowledge graphs let agents actually reason over relationships instead of just retrieving similar chunks. He walks through Cognee&apos;s ECL pipeline (Extract, Cognify, Load) for turning messy multi-source data into something agents can navigate, plus how granular, per-agent access control keeps each agent&apos;s &quot;world model&quot; clean as teams scale.&lt;/p&gt;&lt;p&gt;&lt;/p&gt;&lt;p&gt;🎧 Brought to you by &lt;a rel=&quot;noopener noreferrer nofollow&quot; href=&quot;https://anchorbrowser.io&quot; target=&quot;_blank&quot;&gt;Anchor Browser&lt;/a&gt; — cloud-native browser infrastructure for AI agents.&lt;/p&gt;</itunes:summary><itunes:explicit>no</itunes:explicit><itunes:duration>00:24:17</itunes:duration><itunes:image href="https://hosting-media.riverside.com/media/podcasts/869e72ce-82ce-46e5-9aef-a068c3d4a4b7/episodes/986ac35a-26fe-4fff-84b0-bc271d198b20/images/a3382801-e705-4558-adc8-598a82605ff8.png"/><itunes:season>1</itunes:season><itunes:episode>4</itunes:episode><itunes:title>Episode 004 - Vasilije Markovic, CEO // Cognee</itunes:title><itunes:episodeType>full</itunes:episodeType></item><item><title><![CDATA[Episode 003 - Jean-Denis, CEO // Town AI]]></title><description><![CDATA[<p>What happens when a company stops relying on a shared Notion doc to remember how it works — and lets AI agents do that instead? This week, Gabi is joined by Jean-Denis Greze, co-founder and "Mayor" (not CEO) of Town, an AI assistant startup that just raised a $55M Series A led by Andreessen Horowitz. Before Town, Jean-Denis spent nearly seven years as CTO of Plaid, scaling the engineering org from 20 people to 350 through a $13.4B valuation. He breaks down his thesis of the "self-organizing company," how Town's agents read across email, Slack, and docs without eroding trust between coworkers, and why — despite building an AI assistant for a living — he still won't let his own agent send an email on his behalf.</p><p><b>Guest:</b> Jean-Denis Greze, Co-founder &amp; CEO &amp; "Mayor," Town<br /><b>Links:</b> <a rel="noopener noreferrer nofollow" href="http://town.com" target="_blank">town.com</a></p><p></p><p>🎧 Brought to you by <a rel="noopener noreferrer nofollow" href="https://anchorbrowser.io/" target="_blank"><b>Anchor Browser</b></a> — cloud-native browser infrastructure for AI agents.</p>]]></description><guid isPermaLink="false">9a90647a-a55f-45fd-bf9c-f6d68f998d1d</guid><dc:creator><![CDATA[Anchor Browser]]></dc:creator><pubDate>Mon, 24 Aug 2026 16:15:00 GMT</pubDate><enclosure url="https://api.riverside.com/hosting-analytics/media/eb5cae79e901e601f3ccb72637215938cc774332bccfba80e979c67be5df9cb0/eyJlcGlzb2RlSWQiOiI5YTkwNjQ3YS1hNTVmLTQ1ZmQtYmY5Yy1mNmQ2OGY5OThkMWQiLCJwb2RjYXN0SWQiOiI4NjllNzJjZS04MmNlLTQ2ZTUtOWFlZi1hMDY4YzNkNGE0YjciLCJhY2NvdW50SWQiOiI2OWZjOGNhY2M1MzU3NDUxMTE4NmU5ODciLCJwYXRoIjoibWVkaWEvY2xpcHMvNmE4NzQ1NDYzMzJkMmY2YTg4ZDU4MzZjL2dhYmlzLXN0dWRpby15a2doWS1jb21wb3Nlci0yMDI2LTgtMjBfXzIwLTE5LTUwLm1wMyJ9.mp3" length="73131302" type="audio/mpeg"/><podcast:transcript url="https://hosting-media.riverside.com/media/podcasts/869e72ce-82ce-46e5-9aef-a068c3d4a4b7/episodes/9a90647a-a55f-45fd-bf9c-f6d68f998d1d/transcripts.txt" type="text/plain"/><itunes:summary>&lt;p&gt;What happens when a company stops relying on a shared Notion doc to remember how it works — and lets AI agents do that instead? This week, Gabi is joined by Jean-Denis Greze, co-founder and &quot;Mayor&quot; (not CEO) of Town, an AI assistant startup that just raised a $55M Series A led by Andreessen Horowitz. Before Town, Jean-Denis spent nearly seven years as CTO of Plaid, scaling the engineering org from 20 people to 350 through a $13.4B valuation. He breaks down his thesis of the &quot;self-organizing company,&quot; how Town&apos;s agents read across email, Slack, and docs without eroding trust between coworkers, and why — despite building an AI assistant for a living — he still won&apos;t let his own agent send an email on his behalf.&lt;/p&gt;&lt;p&gt;&lt;b&gt;Guest:&lt;/b&gt; Jean-Denis Greze, Co-founder &amp;amp; CEO &amp;amp; &quot;Mayor,&quot; Town&lt;br /&gt;&lt;b&gt;Links:&lt;/b&gt; &lt;a rel=&quot;noopener noreferrer nofollow&quot; href=&quot;http://town.com&quot; target=&quot;_blank&quot;&gt;town.com&lt;/a&gt;&lt;/p&gt;&lt;p&gt;&lt;/p&gt;&lt;p&gt;🎧 Brought to you by &lt;a rel=&quot;noopener noreferrer nofollow&quot; href=&quot;https://anchorbrowser.io/&quot; target=&quot;_blank&quot;&gt;&lt;b&gt;Anchor Browser&lt;/b&gt;&lt;/a&gt; — cloud-native browser infrastructure for AI agents.&lt;/p&gt;</itunes:summary><itunes:explicit>no</itunes:explicit><itunes:duration>00:38:05</itunes:duration><itunes:image href="https://hosting-media.riverside.com/media/podcasts/869e72ce-82ce-46e5-9aef-a068c3d4a4b7/episodes/9a90647a-a55f-45fd-bf9c-f6d68f998d1d/images/7850f092-8eae-422d-8794-996baae6b2f0.png"/><itunes:season>1</itunes:season><itunes:episode>3</itunes:episode><itunes:title>Episode 003 - Jean-Denis, CEO // Town AI</itunes:title><itunes:episodeType>full</itunes:episodeType></item><item><title><![CDATA[Episode 002 -Peter Holcomb, CEO // Optimo AI]]></title><description><![CDATA[<p>Most companies rolling out AI agents have no idea how much "shadow AI" is already running inside their org. Peter Holcomb, founder &amp; CEO of Optimo AI and fractional CISO (formerly at Datavolo, acquired by Snowflake, and eMed), joins Agent Hour to break down what agent identity and access actually looks like in practice, the compliance gaps that kill enterprise deals, and who's accountable when an agent gets it wrong. Drawing on 100+ engagements helping AI-native companies get through SOC 2, ISO 27001/42001, GDPR, and HIPAA, Peter shares the governance playbook most teams don't build until it's too late.</p><p></p><p>🎧 Brought to you by <a rel="noopener noreferrer nofollow" href="https://anchorbrowser.io" target="_blank">Anchor Browser</a> — cloud-native browser infrastructure for AI agents.</p>]]></description><guid isPermaLink="false">a2cb8863-89ec-4c1d-b6c8-dbe60b375ca6</guid><dc:creator><![CDATA[Anchor Browser]]></dc:creator><pubDate>Mon, 17 Aug 2026 19:42:12 GMT</pubDate><enclosure url="https://api.riverside.com/hosting-analytics/media/b70da57b0a59c72907d694b0cc9614874beca269300a53a20bb73a7fa3400070/eyJlcGlzb2RlSWQiOiJhMmNiODg2My04OWVjLTRjMWQtYjZjOC1kYmU2MGIzNzVjYTYiLCJwb2RjYXN0SWQiOiI4NjllNzJjZS04MmNlLTQ2ZTUtOWFlZi1hMDY4YzNkNGE0YjciLCJhY2NvdW50SWQiOiI2OWZjOGNhY2M1MzU3NDUxMTE4NmU5ODciLCJwYXRoIjoibWVkaWEvY2xpcHMvNmE2YTU4NmVkOTU0MDJiNTA0MzY2MThhL2dhYmlzLXN0dWRpby15a2doWS1jb21wb3Nlci0yMDI2LTctMjlfXzIxLTQ1LTUwLm1wMyJ9.mp3" length="68436784" type="audio/mpeg"/><podcast:transcript url="https://hosting-media.riverside.com/media/podcasts/869e72ce-82ce-46e5-9aef-a068c3d4a4b7/episodes/a2cb8863-89ec-4c1d-b6c8-dbe60b375ca6/transcripts.txt" type="text/plain"/><itunes:summary>&lt;p&gt;Most companies rolling out AI agents have no idea how much &quot;shadow AI&quot; is already running inside their org. Peter Holcomb, founder &amp;amp; CEO of Optimo AI and fractional CISO (formerly at Datavolo, acquired by Snowflake, and eMed), joins Agent Hour to break down what agent identity and access actually looks like in practice, the compliance gaps that kill enterprise deals, and who&apos;s accountable when an agent gets it wrong. Drawing on 100+ engagements helping AI-native companies get through SOC 2, ISO 27001/42001, GDPR, and HIPAA, Peter shares the governance playbook most teams don&apos;t build until it&apos;s too late.&lt;/p&gt;&lt;p&gt;&lt;/p&gt;&lt;p&gt;🎧 Brought to you by &lt;a rel=&quot;noopener noreferrer nofollow&quot; href=&quot;https://anchorbrowser.io&quot; target=&quot;_blank&quot;&gt;Anchor Browser&lt;/a&gt; — cloud-native browser infrastructure for AI agents.&lt;/p&gt;</itunes:summary><itunes:explicit>no</itunes:explicit><itunes:duration>00:35:39</itunes:duration><itunes:image href="https://hosting-media.riverside.com/media/podcasts/869e72ce-82ce-46e5-9aef-a068c3d4a4b7/episodes/a2cb8863-89ec-4c1d-b6c8-dbe60b375ca6/images/349df034-4408-4f45-88f4-8f78d6065fcc.png"/><itunes:season>1</itunes:season><itunes:episode>2</itunes:episode><itunes:title>Episode 002 -Peter Holcomb, CEO // Optimo AI</itunes:title><itunes:episodeType>full</itunes:episodeType></item><item><title><![CDATA[Episode 001 - Antonio Mallia, CEO // Founder Seltz.ai]]></title><description><![CDATA[<p>Antonio Mallia is the founder of Seltz.ai, building search infrastructure from the ground up for a world where the reader is increasingly an agent, not a human. Before Seltz, he spent time at Bloomberg, Amazon's AGI team, and Pinecone, giving him a front-row seat to how retrieval breaks down at scale. In this episode, we dig into why he chose to own the full stack instead of wrapping an existing search index, how trust and authority get scored when there's no human eyeballing a source, and what happens to publishers and the open web as bots increasingly do the reading. We also get his rapid-fire take on what's overrated and underrated in AI search infrastructure today.</p>
]]></description><link>https://podcasters.spotify.com/pod/show/gabi-m-weinberg7/episodes/Episode-001---Antonio-Mallia--CEO--Founder-Seltz-ai-e3mjuk6</link><guid isPermaLink="false">e3b8d010-c465-4a0c-bb46-457e130f89fd</guid><dc:creator><![CDATA[Anchor Browser]]></dc:creator><pubDate>Mon, 27 Jul 2026 14:08:02 GMT</pubDate><enclosure url="https://api.riverside.com/hosting-analytics/media/6b0ac47c10de236c6a49df6a69405e5f42efe6cdc8e0679234979a4ef87640b2/eyJlcGlzb2RlSWQiOiI4OGY0ZjFhYy05ZGM0LTRhNjEtODViZi1hY2QwMDZiN2YwOGUiLCJwb2RjYXN0SWQiOiI4NjllNzJjZS04MmNlLTQ2ZTUtOWFlZi1hMDY4YzNkNGE0YjciLCJhY2NvdW50SWQiOiI2OWZjOGNhY2M1MzU3NDUxMTE4NmU5ODciLCJwYXRoIjoibWVkaWEvaW1wb3J0cy9wb2RjYXN0cy84NjllNzJjZS04MmNlLTQ2ZTUtOWFlZi1hMDY4YzNkNGE0YjcvZXBpc29kZXMvODhmNGYxYWMtOWRjNC00YTYxLTg1YmYtYWNkMDA2YjdmMDhlLzQyODcwMzgzMS00NDEwMC0yLTEyZjM2OTYxNTdjMmYubXAzIn0=.mp3" length="39239679" type="audio/mpeg"/><itunes:summary>&lt;p&gt;Antonio Mallia is the founder of Seltz.ai, building search infrastructure from the ground up for a world where the reader is increasingly an agent, not a human. Before Seltz, he spent time at Bloomberg, Amazon&apos;s AGI team, and Pinecone, giving him a front-row seat to how retrieval breaks down at scale. In this episode, we dig into why he chose to own the full stack instead of wrapping an existing search index, how trust and authority get scored when there&apos;s no human eyeballing a source, and what happens to publishers and the open web as bots increasingly do the reading. We also get his rapid-fire take on what&apos;s overrated and underrated in AI search infrastructure today.&lt;/p&gt;
</itunes:summary><itunes:explicit>no</itunes:explicit><itunes:duration>00:40:52</itunes:duration><itunes:image href="https://hosting-media.riverside.com/media/imports/podcasts/869e72ce-82ce-46e5-9aef-a068c3d4a4b7/episodes/88f4f1ac-9dc4-4a61-85bf-acd006b7f08e/46436792-1785161152430-2cbc30678172d.jpg"/><itunes:season>1</itunes:season><itunes:episode>1</itunes:episode><itunes:title>Episode 001 - Antonio Mallia, CEO // Founder Seltz.ai</itunes:title><itunes:episodeType>full</itunes:episodeType></item></channel></rss>