<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: zihotki</title><link>https://news.ycombinator.com/user?id=zihotki</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 10 Sep 2026 15:22:47 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=zihotki" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by zihotki in "Show HN: Grepathy – Claude made a decision nobody approved"]]></title><description><![CDATA[
<p>Have you tried reading/reviewing the code claude generates? That could have prevented the issue with the accounts.</p>
]]></description><pubDate>Wed, 22 Jul 2026 11:06:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=49004827</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=49004827</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49004827</guid></item><item><title><![CDATA[New comment by zihotki in "Dependabot version updates introduce default package cooldown"]]></title><description><![CDATA[
<p>If everyone starts applying cooldowns, won't it postpone the problem? So now there is a considerable amount of users who are affected and someone from the affected group discovers the infection and reports it.<p>But if everyone will be delaying updates, won't be there less chances to catch it in time? I'm not fully sure if it's possible to preventively scan all NPM packages or how much compute it would require.</p>
]]></description><pubDate>Tue, 14 Jul 2026 22:36:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=48913812</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48913812</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48913812</guid></item><item><title><![CDATA[New comment by zihotki in "YouTrackDB is a general-use object-oriented graph database"]]></title><description><![CDATA[
<p>That's a good question indeed. Also I wonder why they picked java as the implementation language.</p>
]]></description><pubDate>Tue, 14 Jul 2026 05:44:28 +0000</pubDate><link>https://news.ycombinator.com/item?id=48902709</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48902709</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48902709</guid></item><item><title><![CDATA[New comment by zihotki in "Show HN: HTMLDrive – serve HTML files from your Google Drive"]]></title><description><![CDATA[
<p>Since LLM can create the pages for non-techical people, it for sure can also help publishing the page</p>
]]></description><pubDate>Sat, 11 Jul 2026 12:39:49 +0000</pubDate><link>https://news.ycombinator.com/item?id=48871544</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48871544</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48871544</guid></item><item><title><![CDATA[New comment by zihotki in "Dev productivity metrics suck. Ops reviews are key for AI-accelerated eng orgs"]]></title><description><![CDATA[
<p>That sound interesting. Would you mind sharing it?</p>
]]></description><pubDate>Thu, 09 Jul 2026 19:32:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=48851256</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48851256</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48851256</guid></item><item><title><![CDATA[New comment by zihotki in "Why skilled workers come to Germany and then leave again"]]></title><description><![CDATA[
<p>That resonates a lot with my experience in Netherlands. It's way friendlier for expats but the barrier is there</p>
]]></description><pubDate>Tue, 07 Jul 2026 13:07:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=48817242</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48817242</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48817242</guid></item><item><title><![CDATA[New comment by zihotki in "I built a 10 inch mini rack from aluminium extrusions"]]></title><description><![CDATA[
<p>Similar - no (I'd be glad to be proven wrong), but you can find some small suppliers focused on specific niche like <a href="https://www.motedis.com" rel="nofollow">https://www.motedis.com</a> if you need the profiles.</p>
]]></description><pubDate>Wed, 01 Jul 2026 08:29:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=48743790</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48743790</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48743790</guid></item><item><title><![CDATA[New comment by zihotki in "NixOS 26.05"]]></title><description><![CDATA[
<p>I'm building a homelab as well and fully second the ode. It feels incredible to do the setup via llm, like they are the perfect fit.</p>
]]></description><pubDate>Mon, 29 Jun 2026 18:34:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=48723269</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48723269</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48723269</guid></item><item><title><![CDATA[New comment by zihotki in "MaralGPT "Mythos" 9B just released, and this is why I'm proud of my project"]]></title><description><![CDATA[
<p>tldr - Qwen 3.5 9B finetuned on a "simple secret dataset" (500 million tokens of Claude 4.8 Opus, Claude 5 Fable, GPT 5.5 and similar models)</p>
]]></description><pubDate>Mon, 29 Jun 2026 12:02:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=48718139</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48718139</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48718139</guid></item><item><title><![CDATA[New comment by zihotki in "Herdr: Agent multiplexer that lives in your terminal"]]></title><description><![CDATA[
<p>I found that a good tool helps a lot once I switched to Github Copilot app. It solved the friction and mental tax for me. I easily manage 4 sessions in parallel on same or different projects while 2 was max in the past. The bottleneck now is only in review and decision making.</p>
]]></description><pubDate>Mon, 29 Jun 2026 11:57:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=48718093</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48718093</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48718093</guid></item><item><title><![CDATA[New comment by zihotki in "Ask HN: MacBook vs. Dedicated GPU for LLM"]]></title><description><![CDATA[
<p>Not that much - a bit better but still negligible. Prefill is highly parallelizable and benefits from many cores. NVidia GPU's simply have several times more cores than even Pros.</p>
]]></description><pubDate>Sat, 27 Jun 2026 16:58:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=48699828</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48699828</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48699828</guid></item><item><title><![CDATA[New comment by zihotki in "Ask HN: MacBook vs. Dedicated GPU for LLM"]]></title><description><![CDATA[
<p>From personal experience - it works, but you won't get a comfortable time to first token (latency is high). The reason is that prefill on Macs is bad. You need to have a lot more cores to do it quick. It's close to instant for small models on NVidia GPU's but on Macs it takes a few seconds to get the answer for a simple prompt. And the time grows proportionally with your context size.</p>
]]></description><pubDate>Sat, 27 Jun 2026 16:54:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=48699798</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48699798</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48699798</guid></item><item><title><![CDATA[New comment by zihotki in "Show HN: OpenKnowledge – open source AI-first alternative to Obsidian/Notion"]]></title><description><![CDATA[
<p>I wonder how it compares to <a href="https://github.com/refactoringhq/tolaria" rel="nofollow">https://github.com/refactoringhq/tolaria</a> They look similarly scoped. I haven't used it yet and it's great that there is more choice now</p>
]]></description><pubDate>Fri, 26 Jun 2026 09:50:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=48684563</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48684563</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48684563</guid></item><item><title><![CDATA[New comment by zihotki in "Who Does What? Team Topologies for the Agentic Platform"]]></title><description><![CDATA[
<p>In real life only a minority of teams achieved good results according to Circle CI report.<p>And there is one notable anecdotal source proving that it is possible - <a href="https://ideas.fin.ai/p/2x-nine-months-later" rel="nofollow">https://ideas.fin.ai/p/2x-nine-months-later</a></p>
]]></description><pubDate>Tue, 23 Jun 2026 11:20:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=48643296</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48643296</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48643296</guid></item><item><title><![CDATA[New comment by zihotki in "Package Managers need global hooks"]]></title><description><![CDATA[
<p>The solution is called curated package feed. Using private one instead of default public package source, you get a trusted source. All updates should be done by first pulling the new packages to the private feed and only after passing the reviews - update.<p>A little bit of bureaucracy in form of best security practices helps with supply chain attacks.</p>
]]></description><pubDate>Tue, 23 Jun 2026 11:15:28 +0000</pubDate><link>https://news.ycombinator.com/item?id=48643263</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48643263</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48643263</guid></item><item><title><![CDATA[New comment by zihotki in "Stop wasting tokens and re explaining your project between sessions"]]></title><description><![CDATA[
<p>Are there any benchmarks/evals to back the claims? Or how do you know that it helps reducing waste?</p>
]]></description><pubDate>Sun, 21 Jun 2026 22:13:53 +0000</pubDate><link>https://news.ycombinator.com/item?id=48623144</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48623144</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48623144</guid></item><item><title><![CDATA[New comment by zihotki in "What the Fuck Happened to Nerds"]]></title><description><![CDATA[
<p>Isn't that exactly the marketing part? He is the marketing and the channel, he has optimized marketing process.</p>
]]></description><pubDate>Mon, 15 Jun 2026 12:44:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=48540438</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48540438</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48540438</guid></item><item><title><![CDATA[New comment by zihotki in "Microsoft's open source tools were hacked to steal passwords of AI developers"]]></title><description><![CDATA[
<p>And the best recommendation security teams can give - keep your SBOM strict, use min release age policy (sounds more like band-aid). That's a scary world to live in.</p>
]]></description><pubDate>Tue, 09 Jun 2026 08:42:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=48458381</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48458381</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48458381</guid></item><item><title><![CDATA[New comment by zihotki in "Changing how we develop Ladybird"]]></title><description><![CDATA[
<p>I wonder if adding an artificial barrier in form of a donation could help. That's probably the only remaining way to show the good faith.</p>
]]></description><pubDate>Fri, 05 Jun 2026 10:38:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=48410551</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48410551</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48410551</guid></item><item><title><![CDATA[New comment by zihotki in "AI Agent Guidelines for CS336 at Stanford"]]></title><description><![CDATA[
<p>Pencil and paper - that's the classical way a lot of developers from western europe were taught to code. Back to the basics</p>
]]></description><pubDate>Mon, 01 Jun 2026 20:55:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=48362522</link><dc:creator>zihotki</dc:creator><comments>https://news.ycombinator.com/item?id=48362522</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48362522</guid></item></channel></rss>