<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: fesens</title><link>https://news.ycombinator.com/user?id=fesens</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Wed, 29 Apr 2026 16:05:29 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=fesens" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by fesens in "Show HN: Auto-Architecture: Karpathy's Loop, pointed at a CPU"]]></title><description><![CDATA[
<p>His claims are indeed correct; 
Yes, you got my point tks!; 
AND the loop produced architecture gains that are not exclusive to the GoWin FPGA (CoreMark/Mhz is higher than VexRiscV)</p>
]]></description><pubDate>Wed, 29 Apr 2026 15:23:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=47949697</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47949697</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47949697</guid></item><item><title><![CDATA[New comment by fesens in "Show HN: Auto-Architecture: Karpathy's Loop, pointed at a CPU"]]></title><description><![CDATA[
<p>Tks! Did you apply it to hardware design or to another field?</p>
]]></description><pubDate>Wed, 29 Apr 2026 15:19:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=47949665</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47949665</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47949665</guid></item><item><title><![CDATA[New comment by fesens in "Show HN: Auto-Architecture: Karpathy's Loop, pointed at a CPU"]]></title><description><![CDATA[
<p>Whats your take on it? "If I read it, what point should I pay attention to?", I guess is what I'm trying to say</p>
]]></description><pubDate>Wed, 29 Apr 2026 15:18:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=47949649</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47949649</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47949649</guid></item><item><title><![CDATA[New comment by fesens in "Show HN: Auto-Architecture: Karpathy's Loop, pointed at a CPU"]]></title><description><![CDATA[
<p>"Not at all, completely novel idea. What made you thought of such thing?" hahaha</p>
]]></description><pubDate>Wed, 29 Apr 2026 14:17:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=47948814</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47948814</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47948814</guid></item><item><title><![CDATA[New comment by fesens in "Show HN: Auto-Architecture: Karpathy's Loop, pointed at a CPU"]]></title><description><![CDATA[
<p>Yeah, you are totally right. Its a work in progress, and the post was written by an LLM - Im trying to improve on it (dash pun intended).<p>Regarding the benchmark overfitting, absolutely, it's pretty much overfitted. This CPU will only be as good as it benchmark. If I have the time I will try to get some applications and optimize for those.</p>
]]></description><pubDate>Wed, 29 Apr 2026 14:15:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=47948784</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47948784</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47948784</guid></item><item><title><![CDATA[New comment by fesens in "Show HN: Auto-Architecture: Karpathy's Loop, pointed at a CPU"]]></title><description><![CDATA[
<p>I genuinely laughed reading the first words. Yeah, its hard to be novel</p>
]]></description><pubDate>Wed, 29 Apr 2026 14:12:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=47948742</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47948742</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47948742</guid></item><item><title><![CDATA[New comment by fesens in "Show HN: Auto-Architecture: Karpathy's Loop, pointed at a CPU"]]></title><description><![CDATA[
<p>For sure! The hypothesis generation gotta be improved. Your take on the "least likely" is interesting. In the beginning of the repo I was having problems with "hypothesis convergence", your idea may be a nice way to introduce the much needed variability</p>
]]></description><pubDate>Wed, 29 Apr 2026 14:10:27 +0000</pubDate><link>https://news.ycombinator.com/item?id=47948722</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47948722</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47948722</guid></item><item><title><![CDATA[New comment by fesens in "Show HN: Auto-Architecture: Karpathy's Loop, pointed at a CPU"]]></title><description><![CDATA[
<p>Is slop verifiable? If so we can throw it in the loop... The point is that this loop can be pointed at any verifiable work. Yeah you are seeing it raw, the verifier is the principle you talked about. Yes it was fully AI generated, It will be refined</p>
]]></description><pubDate>Wed, 29 Apr 2026 14:07:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=47948686</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47948686</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47948686</guid></item><item><title><![CDATA[New comment by fesens in "Show HN: Auto-Architecture: Karpathy's Loop, pointed at a CPU"]]></title><description><![CDATA[
<p>Nice references! tks</p>
]]></description><pubDate>Wed, 29 Apr 2026 14:04:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=47948641</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47948641</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47948641</guid></item><item><title><![CDATA[New comment by fesens in "Show HN: Auto-Architecture: Karpathy's Loop, pointed at a CPU"]]></title><description><![CDATA[
<p>Absolutely, today it's FPGAs.. Tomorrow can be whole companies</p>
]]></description><pubDate>Wed, 29 Apr 2026 12:46:18 +0000</pubDate><link>https://news.ycombinator.com/item?id=47947561</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47947561</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47947561</guid></item><item><title><![CDATA[New comment by fesens in "Show HN: Auto-Architecture: Karpathy's Loop, pointed at a CPU"]]></title><description><![CDATA[
<p>"I would LOVE somebody to bounce AI off of reversing the architecture and bitstreams for the stupid-ass closed-source FPGAs."<p>The only reason I'm using Gowin is because it has a slightly more mature opensource tooling. Maybe we can apply this loop to nextpnr also</p>
]]></description><pubDate>Wed, 29 Apr 2026 12:45:27 +0000</pubDate><link>https://news.ycombinator.com/item?id=47947554</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47947554</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47947554</guid></item><item><title><![CDATA[New comment by fesens in "Show HN: Auto-Architecture: Karpathy's Loop, pointed at a CPU"]]></title><description><![CDATA[
<p>The frontier is the verifier not in the sense of this project, but to every project. If we have a good verifier for a task, any task, this type of loop can be applied to it. Today LLMs are good enough to tackle FPGA projects, but what this type of loop will be applicable to many more things<p>Board should be arriving next week. I will let you know!</p>
]]></description><pubDate>Wed, 29 Apr 2026 12:44:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=47947536</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47947536</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47947536</guid></item><item><title><![CDATA[New comment by fesens in "Claude.ai is unavailable"]]></title><description><![CDATA[
<p>Ive been receiving rate limits even with full quotas... I guess compute isn't growing as fast as demand</p>
]]></description><pubDate>Tue, 28 Apr 2026 18:37:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=47938586</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47938586</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47938586</guid></item><item><title><![CDATA[New comment by fesens in "Show HN: Ragnerock, an AI data analysis tool"]]></title><description><![CDATA[
<p>I like the way its looking. Maybe because of the familiarity with the Jupyter Notebook</p>
]]></description><pubDate>Tue, 28 Apr 2026 18:31:24 +0000</pubDate><link>https://news.ycombinator.com/item?id=47938488</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47938488</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47938488</guid></item><item><title><![CDATA[Show HN: Auto-Architecture: Karpathy's Loop, pointed at a CPU]]></title><description><![CDATA[
<p>Article URL: <a href="https://github.com/FeSens/auto-arch-tournament/blob/main/docs/auto-arch-tournament-blog-post.md">https://github.com/FeSens/auto-arch-tournament/blob/main/docs/auto-arch-tournament-blog-post.md</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47937380">https://news.ycombinator.com/item?id=47937380</a></p>
<p>Points: 208</p>
<p># Comments: 63</p>
]]></description><pubDate>Tue, 28 Apr 2026 17:12:30 +0000</pubDate><link>https://github.com/FeSens/auto-arch-tournament/blob/main/docs/auto-arch-tournament-blog-post.md</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47937380</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47937380</guid></item><item><title><![CDATA[New comment by fesens in "[dead]"]]></title><description><![CDATA[
<p>Recently I've been noticing that claude "stalls" on a prompt and take minutes to finish, with no apparent work being done. Maybe they are rate limiting, even when you are well under the quotas.</p>
]]></description><pubDate>Mon, 27 Apr 2026 22:34:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=47928293</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47928293</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47928293</guid></item><item><title><![CDATA[New comment by fesens in "[dead]"]]></title><description><![CDATA[
<p>The moment that not releasing a model becomes more financially beneficial than releasing it, that's when AI becomes truly dangerous.
Between doomerism and marketing, the release or rather lack of, Claude Mythos is the first model that really spooked me. Not because of its capabilities, but because of the gatekeeping.
Yes, Anthropic is playing the good guys here, but what if they aren't? What if they gatekeep the next model to Mythos for their own benefit? They would pull so far ahead of society and other software companies, would acquire so much power (through hacking) and money (through arbitrage and trading strategies in the financial markets), that the next model might be the final frontier. The AI 2027 vision doesn't look so distant after all.</p>
]]></description><pubDate>Wed, 08 Apr 2026 12:34:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=47689314</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47689314</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47689314</guid></item><item><title><![CDATA[New comment by fesens in "Container GUI – Native macOS GUI for Apple's Container CLI"]]></title><description><![CDATA[
<p>Apple recently released <a href="https://github.com/apple/container" rel="nofollow">https://github.com/apple/container</a>, a CLI tool for running Linux containers natively on macOS with Apple Silicon. It's fast and lightweight, but it's terminal-only.<p>I built Container GUI — a native SwiftUI app that wraps the CLI and gives you a proper desktop interface for managing containers, images, volumes, networks, and more.<p>Features:
  - Containers: list, inspect, start/stop/kill/delete, live log streaming
  - Images: browse, inspect (config, platform, history), pull with streaming progress
  - Volumes & Networks: full CRUD
  - Builder & System: status, start/stop services, disk usage<p><pre><code>  Tech: Pure SwiftUI + AppKit (for the log viewer), MVVM with @Observable, Swift 6 strict concurrency, zero dependencies. Wraps the CLI via Foundation.Process.
</code></pre>
Install: brew tap FeSens/tap && brew install container-gui<p>Would love feedback!</p>
]]></description><pubDate>Mon, 23 Feb 2026 01:40:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=47116996</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47116996</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47116996</guid></item><item><title><![CDATA[Container GUI – Native macOS GUI for Apple's Container CLI]]></title><description><![CDATA[
<p>Article URL: <a href="https://github.com/FeSens/container-gui">https://github.com/FeSens/container-gui</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47116995">https://news.ycombinator.com/item?id=47116995</a></p>
<p>Points: 3</p>
<p># Comments: 1</p>
]]></description><pubDate>Mon, 23 Feb 2026 01:40:50 +0000</pubDate><link>https://github.com/FeSens/container-gui</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=47116995</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47116995</guid></item><item><title><![CDATA[New comment by fesens in "Opus 4.5 designed a political Magic the Gathering set"]]></title><description><![CDATA[
<p>Its based on the latest political and economic events. I think it may be better than many recent sets.</p>
]]></description><pubDate>Tue, 03 Feb 2026 03:42:49 +0000</pubDate><link>https://news.ycombinator.com/item?id=46866189</link><dc:creator>fesens</dc:creator><comments>https://news.ycombinator.com/item?id=46866189</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46866189</guid></item></channel></rss>