<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: Chyzwar</title><link>https://news.ycombinator.com/user?id=Chyzwar</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Mon, 15 Jun 2026 11:49:51 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=Chyzwar" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by Chyzwar in "Outsourcing plus local AI will soon become more economical vs. frontier labs"]]></title><description><![CDATA[
<p><a href="https://arxiv.org/abs/2604.24827" rel="nofollow">https://arxiv.org/abs/2604.24827</a><p>From this paper</p>
]]></description><pubDate>Tue, 26 May 2026 18:56:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=48284233</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=48284233</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48284233</guid></item><item><title><![CDATA[New comment by Chyzwar in "Outsourcing plus local AI will soon become more economical vs. frontier labs"]]></title><description><![CDATA[
<p>When discussing LLM pricing, people are missing the plot. The subscription token price is 10x-40x cheaper than API pricing. Your 90$ Claude subscriptions give you close to $1000 to $4000 in equivalent API token pricing.<p>The second issue is that the quality of the model “operator” makes a massive difference in the outcomes. Highly skilled senior devs who know how to prompt and have high agency will outperform team people that lack motivation and foundational skills.<p>Lastly, there is a massive difference in capabilities, determinism, and error handling between 5T SOTA models like Opus and tiny distillations from DeepSeek that perform well only in benchmarks.</p>
]]></description><pubDate>Tue, 26 May 2026 14:43:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=48280551</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=48280551</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48280551</guid></item><item><title><![CDATA[New comment by Chyzwar in "Kimi K2.6 just beat Claude, GPT-5.5, and Gemini in a coding challenge"]]></title><description><![CDATA[
<p>You can fix Claude's laziness by modifying the system prompt. <a href="https://gist.github.com/chyzwar/99fe217c3ed336f57c74dcffe371a60b" rel="nofollow">https://gist.github.com/chyzwar/99fe217c3ed336f57c74dcffe371...</a></p>
]]></description><pubDate>Sun, 03 May 2026 10:22:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=47995468</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47995468</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47995468</guid></item><item><title><![CDATA[New comment by Chyzwar in "Claude Code refuses requests or charges extra if your commits mention "OpenClaw""]]></title><description><![CDATA[
<p>Gross margin calculated on API token pricing with discounted training and hardware deprecation.<p>I am no so sure about batch sizes, chatgpt napkin calculation for 5T model show 10-300 sessions.</p>
]]></description><pubDate>Sat, 02 May 2026 05:57:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=47983713</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47983713</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47983713</guid></item><item><title><![CDATA[New comment by Chyzwar in "Claude Code refuses requests or charges extra if your commits mention "OpenClaw""]]></title><description><![CDATA[
<p>All SOTA model providers are losing money. When users run Opus, they are essentially renting a GPU cluster worth half a million dollars for a $100/$200 subscription. If they want to IPO, they need to show a projection toward profit.
For that reason, they want to discourage power users and attract normies.</p>
]]></description><pubDate>Fri, 01 May 2026 11:27:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=47973511</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47973511</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47973511</guid></item><item><title><![CDATA[New comment by Chyzwar in "OpenAI's response to the Axios developer tool compromise"]]></title><description><![CDATA[
<p>Because your fetch most likely mishandle errors, lack retries, fail on redirects and is unnecessary verbose for both people and agents.</p>
]]></description><pubDate>Thu, 23 Apr 2026 10:11:42 +0000</pubDate><link>https://news.ycombinator.com/item?id=47873953</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47873953</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47873953</guid></item><item><title><![CDATA[New comment by Chyzwar in "Ask HN: Building a solo business is impossible?"]]></title><description><![CDATA[
<p>At lot of advice for solo founders  is from people with survivalship bias, vocal sales biz dev and people that had right connections. I think tech solo founder path is different and in today market is very difficult. I can give advice on things that I did wrong so far. Avoid direct stripe integration and complex subscriptions models this would be time sink, use merchant of record and as simple as possible subscription model. Ignore everything non essential for product like usage charts, account management etc. Get decent designer but limit his scope to product features say NO to design system and custom components, use off the shelf UI toolkit with minimal customisation. If you want to raise wait for more traction and avoid startup conferences, better to go for your industry conference instead.</p>
]]></description><pubDate>Mon, 20 Apr 2026 13:41:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=47834191</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47834191</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47834191</guid></item><item><title><![CDATA[New comment by Chyzwar in "What is RISC-V and why it matters to Canonical"]]></title><description><![CDATA[
<p>Exactly without billions in investments this would yet another experimental ISA.</p>
]]></description><pubDate>Sat, 11 Apr 2026 12:10:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=47729891</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47729891</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47729891</guid></item><item><title><![CDATA[New comment by Chyzwar in "GPT 5.4 in practice – Stinks?"]]></title><description><![CDATA[
<p>I recently tested Claude code, opencode, codex on same frontend feature and codex with 5.4 with high effort was the best but most expensive. For me in Europe, Claude Code with 90$ max subscription is the best value for money.<p>My thinking is:<p><pre><code>  codex - best harness
  opencode - best ux/dx
  claude - best value for money</code></pre></p>
]]></description><pubDate>Wed, 08 Apr 2026 07:17:06 +0000</pubDate><link>https://news.ycombinator.com/item?id=47686482</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47686482</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47686482</guid></item><item><title><![CDATA[New comment by Chyzwar in "Post Mortem: axios NPM supply chain compromise"]]></title><description><![CDATA[
<p>This could be controlled by npm. Client ask for available versions anyway. If package is security fix then it can be made available instantly. But this delay gives time for security scanners and time to notify maintainers that package was published.</p>
]]></description><pubDate>Sat, 04 Apr 2026 08:07:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=47636970</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47636970</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47636970</guid></item><item><title><![CDATA[New comment by Chyzwar in "Post Mortem: axios NPM supply chain compromise"]]></title><description><![CDATA[
<p>NPM should fix this mess.<p>Adding postinstall should require approval from NPM.
NPM clients should not install freshly published packages. 
NPM packages should be scanned after publishing.
High profile packages should verify upstream git hash signature.
NPM install should run in sandbox and detect any attempt to install outside project directory.<p>But npm being part of multi trillion company cannot be bothered to fix any of these. Instead they push for tighter integration with GitHub with UX that suck.</p>
]]></description><pubDate>Sat, 04 Apr 2026 07:53:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=47636883</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47636883</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47636883</guid></item><item><title><![CDATA[New comment by Chyzwar in "DRAM pricing is killing the hobbyist SBC market"]]></title><description><![CDATA[
<p>Main producers actually reduced dram output in 2026. When you have few players with very high capital cost you will end up with cartels like light bulb cartel.</p>
]]></description><pubDate>Wed, 01 Apr 2026 23:12:19 +0000</pubDate><link>https://news.ycombinator.com/item?id=47607796</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47607796</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47607796</guid></item><item><title><![CDATA[New comment by Chyzwar in "Axios compromised on NPM – Malicious versions drop remote access trojan"]]></title><description><![CDATA[
<p>Because native fetch lack retries, error handling is verbose, search and body serialization create ton of boilerplate. I use KY http client, small lib on top of fetch with great UX and trusted maintainer.</p>
]]></description><pubDate>Wed, 01 Apr 2026 21:41:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=47606894</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47606894</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47606894</guid></item><item><title><![CDATA[New comment by Chyzwar in "OpenAI closes funding round at an $852B valuation"]]></title><description><![CDATA[
<p>This leak and looking into source code gave me an impulse to try OpenCode with codex models. I am very impressed with how well it works, and the UI is beautiful.</p>
]]></description><pubDate>Wed, 01 Apr 2026 07:26:53 +0000</pubDate><link>https://news.ycombinator.com/item?id=47597933</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47597933</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47597933</guid></item><item><title><![CDATA[New comment by Chyzwar in "404 Deno CEO not found"]]></title><description><![CDATA[
<p>No, they made many wrong architecture decisions that made it fringe project rather than mainstream. You could glimpse on how things could played out by looking into bun.js adoption.</p>
]]></description><pubDate>Sat, 21 Mar 2026 18:02:53 +0000</pubDate><link>https://news.ycombinator.com/item?id=47469518</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47469518</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47469518</guid></item><item><title><![CDATA[New comment by Chyzwar in "Iran war energy shock sparks global push to reduce fossil fuel dependence"]]></title><description><![CDATA[
<p>Russia and Kazakhstan control almost 50% of global production and enrichment of uranium. Even today 17% EU supply come from Russia.<p>Uranium mining is not pretty, read about "in situ leaching" mining.</p>
]]></description><pubDate>Thu, 19 Mar 2026 21:39:06 +0000</pubDate><link>https://news.ycombinator.com/item?id=47446621</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47446621</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47446621</guid></item><item><title><![CDATA[New comment by Chyzwar in "Iran war energy shock sparks global push to reduce fossil fuel dependence"]]></title><description><![CDATA[
<p>Levelized Cost of Energy for solar is 30-60$ and 100-200$ for nuclear. In the case of Spain, it is cheaper to build more energy lines with Morocco and battery storage than to use nuclear. Spain already has some of the cheapest energy in Europe thanks to renewables.<p>In the case of Germany, nuclear makes sense, but it is not clear where you would buy fuel for it, It might still be a supply chain risk since Russia and Kazakhstan are the main players there.</p>
]]></description><pubDate>Thu, 19 Mar 2026 12:57:06 +0000</pubDate><link>https://news.ycombinator.com/item?id=47438601</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47438601</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47438601</guid></item><item><title><![CDATA[New comment by Chyzwar in "A GitHub Issue Title Compromised 4k Developer Machines"]]></title><description><![CDATA[
<p>Both pnpm and yarn implemented 
npmMinimalAgeGate: 1440
enableScripts: false<p>These mostly solve the issue of adding postinstall scripts and packages being compromised.</p>
]]></description><pubDate>Fri, 06 Mar 2026 16:00:19 +0000</pubDate><link>https://news.ycombinator.com/item?id=47276657</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47276657</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47276657</guid></item><item><title><![CDATA[New comment by Chyzwar in "Croatia declared free of landmines after 31 years"]]></title><description><![CDATA[
<p>In conflict between equals, landmines are the only practical way to restrict the mobility of the enemy. That's why 20% of Ukraine is contaminated by mines. If you were official and your choices would be losing and more people dying or placing more landmines that can be cleared over 20 years, what would you do?</p>
]]></description><pubDate>Sat, 28 Feb 2026 09:21:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=47192740</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47192740</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47192740</guid></item><item><title><![CDATA[New comment by Chyzwar in "JavaScript-heavy approaches are not compatible with long-term performance goals"]]></title><description><![CDATA[
<p>As someone that spend many days on performance I will tell you that bundle size have minimal impact on performance. In most cases backend dominate any performance discussion.<p>Apps are slow because caching is missing, naive pagination, poorly written API calls waterfall, missing db indexes, bad data model, database choice or slow serverless environment. Extra 1MB of bundle add maybe 100ms to one time initial load of app that can mitigated trivially with code splitting.</p>
]]></description><pubDate>Mon, 16 Feb 2026 11:32:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=47033802</link><dc:creator>Chyzwar</dc:creator><comments>https://news.ycombinator.com/item?id=47033802</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47033802</guid></item></channel></rss>