<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: adi_pradhan</title><link>https://news.ycombinator.com/user?id=adi_pradhan</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Mon, 15 Jun 2026 10:51:14 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=adi_pradhan" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by adi_pradhan in "DeepSeek makes the V4 Pro price discount permanent"]]></title><description><![CDATA[
<p>Great headline cost reduction, but has anyone here actually used the API in production?<p>I'm constantly getting provider not available at least when using the DeepSeek provider for DeepSeek v4 flash or pro through Open Router.<p>It seems like there isn't enough capacity to actually serve production traffic</p>
]]></description><pubDate>Sun, 24 May 2026 15:53:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=48258288</link><dc:creator>adi_pradhan</dc:creator><comments>https://news.ycombinator.com/item?id=48258288</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48258288</guid></item><item><title><![CDATA[New comment by adi_pradhan in "LinkedIn uses 2.4 GB RAM across two tabs"]]></title><description><![CDATA[
<p>LOL you've nailed LinkedinSpeak here</p>
]]></description><pubDate>Sun, 29 Mar 2026 15:40:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=47564099</link><dc:creator>adi_pradhan</dc:creator><comments>https://news.ycombinator.com/item?id=47564099</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47564099</guid></item><item><title><![CDATA[New comment by adi_pradhan in "Prompt-caching – auto-injects Anthropic cache breakpoints (90% token savings)"]]></title><description><![CDATA[
<p>This is applicable only to the API from what i understand. Since claude code already caches quite aggressively (try npx ccusage)<p>Also the anthropic API did already introduce prompt-caching
<a href="https://platform.claude.com/docs/en/build-with-claude/prompt-caching" rel="nofollow">https://platform.claude.com/docs/en/build-with-claude/prompt...</a><p>What is new here?</p>
]]></description><pubDate>Fri, 13 Mar 2026 12:56:53 +0000</pubDate><link>https://news.ycombinator.com/item?id=47363873</link><dc:creator>adi_pradhan</dc:creator><comments>https://news.ycombinator.com/item?id=47363873</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47363873</guid></item><item><title><![CDATA[New comment by adi_pradhan in "Show HN: Tailwind Template Directory"]]></title><description><![CDATA[
<p>Same. 3.5 Sonnet seems particularly good at React and Tailwind. I'm moving away from other frameworks and libraries to React and Tailwind because of this.</p>
]]></description><pubDate>Wed, 10 Jul 2024 16:57:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=40928905</link><dc:creator>adi_pradhan</dc:creator><comments>https://news.ycombinator.com/item?id=40928905</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40928905</guid></item><item><title><![CDATA[New comment by adi_pradhan in "Video generation models as world simulators"]]></title><description><![CDATA[
<p>Thanks for adding the specific case. I think with testing these sort of limited domain applications make sense.<p>It'll be much harder for more open ended world problems where the physics encountered may be rare enough in the dataset that the simulation breaks unexpectedly. For example a glass smashing into the floor. The model doesn't simulate that causally afaik</p>
]]></description><pubDate>Fri, 16 Feb 2024 02:20:53 +0000</pubDate><link>https://news.ycombinator.com/item?id=39392326</link><dc:creator>adi_pradhan</dc:creator><comments>https://news.ycombinator.com/item?id=39392326</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39392326</guid></item></channel></rss>