<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: vochsel</title><link>https://news.ycombinator.com/user?id=vochsel</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Mon, 06 Apr 2026 02:40:35 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=vochsel" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by vochsel in "What if the browser built the UI for you?"]]></title><description><![CDATA[
<p>maybe one day our llm browser ui will remove those too ;)</p>
]]></description><pubDate>Sun, 05 Apr 2026 05:35:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=47646402</link><dc:creator>vochsel</dc:creator><comments>https://news.ycombinator.com/item?id=47646402</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47646402</guid></item><item><title><![CDATA[New comment by vochsel in "What if the browser built the UI for you?"]]></title><description><![CDATA[
<p>I do like this idea, and agree on the timelines of the world grappling with what to change and what to keep with these new capabilities.<p>Having a traditional web page with styles and assets AND the spec allow LLM's to be a bit more guided by the original site's design and intent. More of a remix or like Arc's boosts/skills feature.<p>There's also the reality the a lot of the things you'd want to be promptable (sorting, functionality, enrichment) couldn't be done on just the front end. You need some mix of UI and API logic to be promptable...</p>
]]></description><pubDate>Sun, 05 Apr 2026 05:25:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=47646349</link><dc:creator>vochsel</dc:creator><comments>https://news.ycombinator.com/item?id=47646349</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47646349</guid></item><item><title><![CDATA[New comment by vochsel in "OpenScreen is an open-source alternative to Screen Studio"]]></title><description><![CDATA[
<p>+1 this, I bought it when it was a flat fee for a year of updates. 
That year's up now and I'd happily pay for another, but not as a subscription...</p>
]]></description><pubDate>Sun, 05 Apr 2026 05:19:45 +0000</pubDate><link>https://news.ycombinator.com/item?id=47646328</link><dc:creator>vochsel</dc:creator><comments>https://news.ycombinator.com/item?id=47646328</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47646328</guid></item><item><title><![CDATA[Show HN: Walkie-Talkie – Access your terminal from any browser with one command]]></title><description><![CDATA[
<p>I've spent the last month building an electron agent orchestrator wrapper. But it occurred to me you could also go the other way - expose my terminal via local web sockets to a website, with any sort of ui...</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47540778">https://news.ycombinator.com/item?id=47540778</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Fri, 27 Mar 2026 09:59:56 +0000</pubDate><link>https://www.walkie-talkie.dev/</link><dc:creator>vochsel</dc:creator><comments>https://news.ycombinator.com/item?id=47540778</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47540778</guid></item><item><title><![CDATA[New comment by vochsel in "Show HN: Dump – easily share context with AI"]]></title><description><![CDATA[
<p>thanks!</p>
]]></description><pubDate>Wed, 18 Mar 2026 10:59:24 +0000</pubDate><link>https://news.ycombinator.com/item?id=47424050</link><dc:creator>vochsel</dc:creator><comments>https://news.ycombinator.com/item?id=47424050</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47424050</guid></item><item><title><![CDATA[Show HN: Dump – easily share context with AI]]></title><description><![CDATA[
<p>I wanted a very simple way to dump prompts, links, and todo lists into my claude code and codex agents.<p>And to work outside of the context window. This works particularly well for sharing "Projects" between Claude/ChatGPT etc.<p>It's open source here; <a href="https://github.com/Vochsel/dump.page" rel="nofollow">https://github.com/Vochsel/dump.page</a><p>Anything you dump on the board becomes an llms.txt - spatially sorted implicitly and explicity sorted via connection edges.<p>Would love HN's thoughts!</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47421093">https://news.ycombinator.com/item?id=47421093</a></p>
<p>Points: 7</p>
<p># Comments: 2</p>
]]></description><pubDate>Wed, 18 Mar 2026 02:54:08 +0000</pubDate><link>https://www.dump.page</link><dc:creator>vochsel</dc:creator><comments>https://news.ycombinator.com/item?id=47421093</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47421093</guid></item><item><title><![CDATA[New comment by vochsel in "Show HN: I made a daily Hacker News AI podcast in Magpai"]]></title><description><![CDATA[
<p>Magpai podcasts are an easily prompt-able way of creating a single host podcast. Using @ style references, you can tag any of your linked documents or data sources in Magpai.<p>I'm adding the ability to share podcasts/magic-documents made in Magpai soon, so this also doubles as a stress test :)</p>
]]></description><pubDate>Tue, 01 Oct 2024 04:42:41 +0000</pubDate><link>https://news.ycombinator.com/item?id=41704664</link><dc:creator>vochsel</dc:creator><comments>https://news.ycombinator.com/item?id=41704664</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41704664</guid></item><item><title><![CDATA[Show HN: I made a daily Hacker News AI podcast in Magpai]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.magpai.app/share/n7R91q">https://www.magpai.app/share/n7R91q</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=41704663">https://news.ycombinator.com/item?id=41704663</a></p>
<p>Points: 3</p>
<p># Comments: 1</p>
]]></description><pubDate>Tue, 01 Oct 2024 04:42:41 +0000</pubDate><link>https://www.magpai.app/share/n7R91q</link><dc:creator>vochsel</dc:creator><comments>https://news.ycombinator.com/item?id=41704663</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41704663</guid></item><item><title><![CDATA[New comment by vochsel in "NotebookLM's automatically generated podcasts are surprisingly effective"]]></title><description><![CDATA[
<p>They've really nailed the back and fourth of the two speakers!<p>It would be interesting to know if it's multimodal voice, or just clever prompting and recombining...<p>I added single voice podcasts to Magpai after seeing how useful this was. Allows for a bit more customisation of the podcast too <a href="https://www.youtube.com/watch?v=OEsh9MlbA6s" rel="nofollow">https://www.youtube.com/watch?v=OEsh9MlbA6s</a><p>I've got a daily podcast of hackernews being generated here too: <a href="https://www.magpai.app/share/n7R91q" rel="nofollow">https://www.magpai.app/share/n7R91q</a></p>
]]></description><pubDate>Mon, 30 Sep 2024 05:43:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=41693924</link><dc:creator>vochsel</dc:creator><comments>https://news.ycombinator.com/item?id=41693924</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41693924</guid></item></channel></rss>