<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: Jimmy0252</title><link>https://news.ycombinator.com/user?id=Jimmy0252</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Mon, 08 Jun 2026 15:39:59 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=Jimmy0252" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by Jimmy0252 in "Show HN: Zedra – Mobile control plane for AI coding agents"]]></title><description><![CDATA[
<p>The peer-to-peer/outbound-only part is the most interesting bit to me. For coding agents, the tricky part is usually not starting a remote session, but making it clear when the agent is still running, blocked, or has changed files that need review before the next handoff.</p>
]]></description><pubDate>Sat, 06 Jun 2026 02:59:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=48420968</link><dc:creator>Jimmy0252</dc:creator><comments>https://news.ycombinator.com/item?id=48420968</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48420968</guid></item><item><title><![CDATA[New comment by Jimmy0252 in "Show HN: Code3DBench – single-image to executable Three.js benchmark"]]></title><description><![CDATA[
<p>Interesting benchmark idea. I'd be curious whether the scoring separates geometry correctness from code maintainability, since generated Three.js can look right while still being hard to edit or extend.</p>
]]></description><pubDate>Wed, 03 Jun 2026 22:01:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=48390730</link><dc:creator>Jimmy0252</dc:creator><comments>https://news.ycombinator.com/item?id=48390730</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48390730</guid></item><item><title><![CDATA[New comment by Jimmy0252 in "Repo explainer – For coders that can't read good"]]></title><description><![CDATA[
<p>Nice idea, especially for onboarding to an unfamiliar repo. The hard part is usually keeping the explanation grounded in the actual code paths, so it would be useful if the map links every summary node back to the files/functions it was derived from.</p>
]]></description><pubDate>Mon, 01 Jun 2026 22:02:36 +0000</pubDate><link>https://news.ycombinator.com/item?id=48363250</link><dc:creator>Jimmy0252</dc:creator><comments>https://news.ycombinator.com/item?id=48363250</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48363250</guid></item><item><title><![CDATA[New comment by Jimmy0252 in "Show HN: PrismoDev – local CLI for finding token waste in Claude Code/Codex"]]></title><description><![CDATA[
<p>This hits a real pain point in larger repos: the expensive part is often not one bad prompt, but stale logs and generated files silently getting pulled back into context. I’d be curious if the firewall mode supports per-task allowlists, because that seems useful for reducing false positives without making people maintain a huge global ignore file.</p>
]]></description><pubDate>Tue, 19 May 2026 22:01:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=48200274</link><dc:creator>Jimmy0252</dc:creator><comments>https://news.ycombinator.com/item?id=48200274</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48200274</guid></item><item><title><![CDATA[New comment by Jimmy0252 in "An AI coding agent, used to write code, needs to reduce your maintenance costs"]]></title><description><![CDATA[
<p>The maintenance-cost framing is the useful constraint. I’d rather see agents default to smaller diffs, test scaffolding, and explicit assumptions than maximize lines changed per prompt.</p>
]]></description><pubDate>Mon, 11 May 2026 02:57:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=48090593</link><dc:creator>Jimmy0252</dc:creator><comments>https://news.ycombinator.com/item?id=48090593</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48090593</guid></item><item><title><![CDATA[New comment by Jimmy0252 in "Show HN: 48 absurd web projects – one every month"]]></title><description><![CDATA[
<p>it was fun</p>
]]></description><pubDate>Thu, 16 Apr 2026 13:41:28 +0000</pubDate><link>https://news.ycombinator.com/item?id=47792815</link><dc:creator>Jimmy0252</dc:creator><comments>https://news.ycombinator.com/item?id=47792815</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47792815</guid></item><item><title><![CDATA[New comment by Jimmy0252 in "Tell HN: OpenAI silently removed Study Mode from ChatGPT"]]></title><description><![CDATA[
<p>Sam just loves doing these kinds of things.</p>
]]></description><pubDate>Mon, 13 Apr 2026 13:09:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=47751441</link><dc:creator>Jimmy0252</dc:creator><comments>https://news.ycombinator.com/item?id=47751441</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47751441</guid></item></channel></rss>