<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: ditchfieldcaleb</title><link>https://news.ycombinator.com/user?id=ditchfieldcaleb</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Mon, 15 Jun 2026 12:20:50 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=ditchfieldcaleb" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by ditchfieldcaleb in "Ask HN: What are you working on? (May 2026)"]]></title><description><![CDATA[
<p>Oh, this is cute and a great first game! I'm working on my first game as well (a top-down 2D tower defense game).<p>What engine or framework did you end up going with? I looked into Unity, tried Godot for a few weeks, but landed on just making a Typescript-powered canvas game with PixiJS for graphics rendering. Found it <i>much</i> easier doing it this way instead of having to learn a game engine.</p>
]]></description><pubDate>Mon, 11 May 2026 01:45:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=48090156</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=48090156</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48090156</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "Vibe coding and agentic engineering are getting closer than I'd like"]]></title><description><![CDATA[
<p>Say what now</p>
]]></description><pubDate>Thu, 07 May 2026 03:01:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=48044946</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=48044946</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48044946</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "Vibe coding and agentic engineering are getting closer than I'd like"]]></title><description><![CDATA[
<p>I agree with you on everything you said here except:<p>> when you know how the thing works and have that mental context, you will always be faster than an AI<p>That's just plain false, honestly. No one can type at the speed AI can code, even factoring in the time you need to spend to properly write out the spec & design rules the AI needs to follow when implementing your app/feature/whatever. And that gap will only increase as LLMs get more intelligent.</p>
]]></description><pubDate>Thu, 07 May 2026 02:58:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=48044933</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=48044933</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48044933</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "Codex's precision and attention to detail is *crazy* when set up correctly"]]></title><description><![CDATA[
<p>Oh yeah, I did a deep-dive into neural networks (both artificial and human) for vision processing, it's super dope stuff. The human vision processing system is remarkably similar to some of the AI stuff we've built for image processing!</p>
]]></description><pubDate>Wed, 06 May 2026 15:24:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=48037327</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=48037327</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48037327</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "Agent Skills"]]></title><description><![CDATA[
<p>TBH this alone is worth $20 (but don't tell OpenAI that).</p>
]]></description><pubDate>Wed, 06 May 2026 04:21:20 +0000</pubDate><link>https://news.ycombinator.com/item?id=48032135</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=48032135</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48032135</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "Agent Skills"]]></title><description><![CDATA[
<p>> I like thinking, solving problems and typing out code myself.<p>I get this, I totally do, and I kind of hate relegating myself to doing "project manager" work instead of "software engineer" work, but the productivity gains make it no contest on whether to use AI here. Once I comprehensively validate the spec for a new feature, Codex just one-shots it basically every time. I'm talking thousands of lines of code in a single 3-hour session, with much of my time being spent browsing the internet while I wait for Codex to run in 15-20 minute sessions.<p>I'd estimate at <i>least</i> a 20x speedup in my ability to ship.<p>(and before you say it, yes, I review every single line of code before merging anything, so no - it's not AI slop)</p>
]]></description><pubDate>Wed, 06 May 2026 04:15:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=48032102</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=48032102</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48032102</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "Codex's precision and attention to detail is *crazy* when set up correctly"]]></title><description><![CDATA[
<p>I was wrong & you called me out on it, not rude, all good.<p>My first software job out of college was actually a QA Automation / SDET position, wrote an automated framework with Ruby + Selenium + Browserstack which did take screenshots of the app, but the app loaded dynamic content and there were frequent feature adjustments so no two screenshots were ever identical.<p>All other jobs I've had since then have been writing smart contracts for Ethereum apps - 100% backend, (I <i>hate</i> having to deal with frontend) so all our tests were just units & coverage & what have you.<p>I suppose if your environment holds constant and your features don't change frontend structure or behavior (eg refactors), then this is what you should expect.<p>Though, do note that this only works because my app is based on a tick/game-loop system <i>without</i> callbacks; if this was the standard game-development pattern of callbacks & message handling (especially w/ React / JS) to invoke events, it wouldn't work, because the timing would be <i>slightly</i> different each time, and an enemy would be a <i>few</i> pixels to the left/right of its position in the past run.</p>
]]></description><pubDate>Wed, 06 May 2026 02:26:06 +0000</pubDate><link>https://news.ycombinator.com/item?id=48031403</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=48031403</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48031403</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "Codex's precision and attention to detail is *crazy* when set up correctly"]]></title><description><![CDATA[
<p>Huh! Well, I stand corrected. I've never seen that done (but I've only worked at startups with < 20 headcount for my entire software career so far, so that might be why).</p>
]]></description><pubDate>Wed, 06 May 2026 02:03:22 +0000</pubDate><link>https://news.ycombinator.com/item?id=48031258</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=48031258</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48031258</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "Codex's precision and attention to detail is *crazy* when set up correctly"]]></title><description><![CDATA[
<p>...yes, I'm aware of what a checksum is.</p>
]]></description><pubDate>Wed, 06 May 2026 01:48:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=48031174</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=48031174</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48031174</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "Codex's precision and attention to detail is *crazy* when set up correctly"]]></title><description><![CDATA[
<p>Not sure if you're getting what I'm saying.<p>No one that I've seen takes automated screenshots of webapps or games or what have you at pre-determined timestamps <i>to make sure the app looks pixel-identical</i> with every change.<p>(regardless of the method; the SHA'ing isn't the point here, the point is that it's a shortcut instead of "inspect the image for any regressions", since we don't <i>need</i> to inspect the image at all if it is identical)</p>
]]></description><pubDate>Wed, 06 May 2026 01:47:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=48031165</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=48031165</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48031165</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "Codex's precision and attention to detail is *crazy* when set up correctly"]]></title><description><![CDATA[
<p>Sure, it's been done before, and I'm sure not just limited to SGI, but <i>no one</i> does this for regular apps these days - never heard of it before. I just find it neat that Codex came up with this - not something I ever would have.<p>Edit: I'm not saying no one does checksums to compare files (lol). I'm saying no one takes screenshots at specific timestamps within an app or game's lifecycle and then compares them to ensure they're identical.<p>Edit 2: Whoops, looks like I'm wrong and this is apparently a pretty common thing (but not at the startups I've worked at, /shrug). I still think it's cool that Codex is doing it without being told to, though.</p>
]]></description><pubDate>Wed, 06 May 2026 00:12:48 +0000</pubDate><link>https://news.ycombinator.com/item?id=48030520</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=48030520</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48030520</guid></item><item><title><![CDATA[Codex's precision and attention to detail is *crazy* when set up correctly]]></title><description><![CDATA[
<p>Lately I've been working on a Tower Defense game with Codex, in part to learn how game development works and in part to see how far I can get using <i>just</i> Codex, no manual coding at all. I've got my AGENTS md & my CODESTYLE md & six other ALLCAPS md files etc, and am working on some refactoring to keep the codebase clean & file sizes low, etc.<p>And then I see this in the ExecPlan for my latest refactor:<p>---<p># Observations<p>- Observation: The refactor made the screenshots pixel-identical after the baseline was recaptured correctly.<p>Evidence: sha256sum screenshots/before-implementation-x.png screenshots/after-implementation-x.png reported matching hashes for before/after pairs 1, 2, and 3.<p>---<p>Which is crazy! I've never told Codex to do an <i>sha compare</i> on before/after screenshots of the app, but I do have instructions in my PLANS.md to take before & after screenshots of the webapp for the game to make sure we avoid frontend regressions (it uses GPT-Image-2 for analysis). So for non-frontend impacting changes, of course nothing should be different between screenshots taken at <i>identical timestamps</i> into the game start.<p>But doing an explicit SHA compare - that's just...not something I would've <i>ever</i> thought of. Wild.</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=48030302">https://news.ycombinator.com/item?id=48030302</a></p>
<p>Points: 5</p>
<p># Comments: 13</p>
]]></description><pubDate>Tue, 05 May 2026 23:45:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=48030302</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=48030302</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48030302</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "SpåraKoff"]]></title><description><![CDATA[
<p>> the SpåraKoff's seats are not comfortable, but its scheduled route covers most of Helsinki's usual sights, its drinks are reasonably priced, and there is no irritating commentary<p>No commentary? I'm in.</p>
]]></description><pubDate>Sun, 23 Jul 2023 19:53:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=36839114</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=36839114</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=36839114</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "ChatGPT can now call Wolfram Alpha"]]></title><description><![CDATA[
<p>Wow. This prompt is incredibly well engineered.</p>
]]></description><pubDate>Thu, 23 Mar 2023 19:24:06 +0000</pubDate><link>https://news.ycombinator.com/item?id=35280007</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=35280007</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=35280007</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "7 Weeks AI Alignment Curriculum"]]></title><description><![CDATA[
<p>You're right - the odds of success are incredibly slim. But we may only have one shot, because the first mid-air collision in this analogy could literally be the end of the world.</p>
]]></description><pubDate>Sun, 05 Mar 2023 18:00:22 +0000</pubDate><link>https://news.ycombinator.com/item?id=35032134</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=35032134</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=35032134</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "High-res image reconstruction with latent diffusion models from human brain"]]></title><description><![CDATA[
<p>Ah. This makes more sense. Thanks.</p>
]]></description><pubDate>Fri, 03 Mar 2023 23:23:20 +0000</pubDate><link>https://news.ycombinator.com/item?id=35016362</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=35016362</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=35016362</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "High-res image reconstruction with latent diffusion models from human brain"]]></title><description><![CDATA[
<p>What are you talking about? They didn't train a model for this. That's why it's so impressive.</p>
]]></description><pubDate>Fri, 03 Mar 2023 21:26:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=35015243</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=35015243</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=35015243</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "Ask HN: How long until AI shows general intelligence without failures?"]]></title><description><![CDATA[
<p>Not the brightest, but a valiant attempt for a lower-iq...entity, I'd say.</p>
]]></description><pubDate>Sun, 05 Feb 2023 20:27:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=34668957</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=34668957</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=34668957</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "Ask HN: How long until AI shows general intelligence without failures?"]]></title><description><![CDATA[
<p>FWIW I agree with you, but was attempting to provoke the discussion of "what is intelligence" before strictly taking that opposing stance to the parent comment.</p>
]]></description><pubDate>Sun, 05 Feb 2023 14:16:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=34665048</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=34665048</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=34665048</guid></item><item><title><![CDATA[New comment by ditchfieldcaleb in "Ask HN: How long until AI shows general intelligence without failures?"]]></title><description><![CDATA[
<p>I was really on your side here when I tried this...until I was one up'd.<p>> What is the last business day of the month?<p>The last business day of the month is typically the last weekday of the month (excluding weekends). For example, the last business day of August 2020 would be Friday, August 28th.<p>> But August has 31 days, and August 31st, 2020 is a Monday.<p>In this case, the last business day of the month would be Friday, August 28th.<p>> Why not Monday, the 31st?<p>Monday, August 31st, 2020 is a public holiday in some countries, so it would not be considered a business day. Therefore, the last business day of the month would be the previous Friday, August 28th.</p>
]]></description><pubDate>Sun, 05 Feb 2023 06:15:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=34662452</link><dc:creator>ditchfieldcaleb</dc:creator><comments>https://news.ycombinator.com/item?id=34662452</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=34662452</guid></item></channel></rss>