<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: purplemantis</title><link>https://news.ycombinator.com/user?id=purplemantis</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Wed, 18 Nov 2026 03:40:43 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=purplemantis" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by purplemantis in "We Tested Coding Agent Harnesses. Thinking Harder Beat All of Them"]]></title><description><![CDATA[
<p>I wonder how gstack stacks up (Garry Tan's special coding setup, which he spends $200K/yr on: <a href="https://techcrunch.com/2026/11/02/y-combinators-garry-tan-uses-ai-agents-so-much-he-spends-up-to-200k-a-year-on-tokens/" rel="nofollow">https://techcrunch.com/2026/11/02/y-combinators-garry-tan-us...</a>)<p>Maybe all the extra features and scaffolding helped when he first started building it... But I'd bet that raw Claude Code and Codex are smart enough now that they make things worse.</p>
]]></description><pubDate>Wed, 04 Nov 2026 21:03:49 +0000</pubDate><link>https://news.ycombinator.com/item?id=50371491</link><dc:creator>purplemantis</dc:creator><comments>https://news.ycombinator.com/item?id=50371491</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50371491</guid></item><item><title><![CDATA[New comment by purplemantis in "Ripple-65B Preview: U.S. Open-Weight SOTA on Agentic Coding"]]></title><description><![CDATA[
<p>These are pretty odd benchmarks to use. Terminal Bench 2.1 is old. SWE-bench was disavowed by OpenAI. DeepSWE is clearly benchmark-hacked by Gemini.</p>
]]></description><pubDate>Wed, 04 Nov 2026 20:56:53 +0000</pubDate><link>https://news.ycombinator.com/item?id=50371398</link><dc:creator>purplemantis</dc:creator><comments>https://news.ycombinator.com/item?id=50371398</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50371398</guid></item><item><title><![CDATA[New comment by purplemantis in "North Korean fake job interview led to $6.3B Binance heist"]]></title><description><![CDATA[
<p>“shared a social media account with him, including personal photos and videos posted on the platform, in order to build trust" -- I wonder if these were real photos and videos, or AI fakes. It's crazy how many fake profiles Instagram pushes into my feed lately, and how many people probably don't even realize they're fake.</p>
]]></description><pubDate>Wed, 04 Nov 2026 19:25:36 +0000</pubDate><link>https://news.ycombinator.com/item?id=50370333</link><dc:creator>purplemantis</dc:creator><comments>https://news.ycombinator.com/item?id=50370333</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50370333</guid></item><item><title><![CDATA[New comment by purplemantis in "Tell HN: Gemini trains on chat history, even for subscribers (personal accounts)"]]></title><description><![CDATA[
<p>It's stuff like this that makes me happy that at least one big tech company (Apple) cares about privacy. Hope they don't lose that. Although I wonder how much of that is exactly why they're behind in AI...</p>
]]></description><pubDate>Wed, 04 Nov 2026 18:26:36 +0000</pubDate><link>https://news.ycombinator.com/item?id=50369565</link><dc:creator>purplemantis</dc:creator><comments>https://news.ycombinator.com/item?id=50369565</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50369565</guid></item><item><title><![CDATA[New comment by purplemantis in "Kimi K3.1 Pro model weights are out"]]></title><description><![CDATA[
<p>3T parameters! I've been really impressed with Kimi's coding performance. It's replaced a lot of my Claude Code workflows, which just got way too expensive. Wonder how big Fable and Opus are in comparison.</p>
]]></description><pubDate>Wed, 04 Nov 2026 18:24:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=50369531</link><dc:creator>purplemantis</dc:creator><comments>https://news.ycombinator.com/item?id=50369531</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50369531</guid></item><item><title><![CDATA[iOS 26 Review: Through a glass, liquidly]]></title><description><![CDATA[
<p>Article URL: <a href="https://sixcolors.com/post/2025/09/ios-26-review-through-a-glass-liquidly/">https://sixcolors.com/post/2025/09/ios-26-review-through-a-glass-liquidly/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=45328501">https://news.ycombinator.com/item?id=45328501</a></p>
<p>Points: 4</p>
<p># Comments: 0</p>
]]></description><pubDate>Mon, 22 Sep 2025 02:16:58 +0000</pubDate><link>https://sixcolors.com/post/2025/09/ios-26-review-through-a-glass-liquidly/</link><dc:creator>purplemantis</dc:creator><comments>https://news.ycombinator.com/item?id=45328501</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45328501</guid></item><item><title><![CDATA[Electric Mercedes-Benz drove 747 miles without recharging, 2x Tesla's Model S]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.businessinsider.com/mercedes-electric-car-longest-range-tesla-model-s-eqs-eqxx-2022-6">https://www.businessinsider.com/mercedes-electric-car-longest-range-tesla-model-s-eqs-eqxx-2022-6</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=31884946">https://news.ycombinator.com/item?id=31884946</a></p>
<p>Points: 28</p>
<p># Comments: 45</p>
]]></description><pubDate>Sun, 26 Jun 2022 15:39:26 +0000</pubDate><link>https://www.businessinsider.com/mercedes-electric-car-longest-range-tesla-model-s-eqs-eqxx-2022-6</link><dc:creator>purplemantis</dc:creator><comments>https://news.ycombinator.com/item?id=31884946</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=31884946</guid></item></channel></rss>