<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: prettyblocks</title><link>https://news.ycombinator.com/user?id=prettyblocks</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Mon, 25 May 2026 00:03:11 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=prettyblocks" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by prettyblocks in "The last six months in LLMs in five minutes"]]></title><description><![CDATA[
<p>In my experience most people with the type of critique I'm seeing from you <i>have</i> only tried it one time or have not taken the time to invest in an environment/process that will work for agentic coding.<p>My question is not so much about sharing a cherry picked example, but the question was more like "have you tried in earnest to make it work". That's the part that wasn't clear from your original post. But you say you have, and you weren't impressed. Fair enough. I'm not trying to convince you otherwise, but I encourage people to give the tools a fair chance before throwing up their hands and deciding it's meh.<p>Having said all that, you're right there isn't a shared experience.</p>
]]></description><pubDate>Tue, 19 May 2026 14:53:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=48194139</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=48194139</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48194139</guid></item><item><title><![CDATA[New comment by prettyblocks in "The last six months in LLMs in five minutes"]]></title><description><![CDATA[
<p>I'm curious. What have you actually tried? Are you just prompting the LLM with one off tasks? For good results, you need to take the time to read the documentation for the harness you are using and configure your environment. This tuning can take dozens hours to nail down. Then there's the actual approach for working on your projects. Many people that have good results with agentic coding actually spend the bulk of their time in plan mode where they go back and forth with the LLM designing a granular playbook for the task at hand before they ever have it write any code.</p>
]]></description><pubDate>Tue, 19 May 2026 13:39:49 +0000</pubDate><link>https://news.ycombinator.com/item?id=48193191</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=48193191</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48193191</guid></item><item><title><![CDATA[New comment by prettyblocks in "The American Rebellion Against AI Is Gaining Steam"]]></title><description><![CDATA[
<p>#2 is not really an option though. It's more like #1 or #3.</p>
]]></description><pubDate>Tue, 19 May 2026 03:14:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=48188765</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=48188765</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48188765</guid></item><item><title><![CDATA[New comment by prettyblocks in "Running local models on an M4 with 24GB memory"]]></title><description><![CDATA[
<p>If you use lmstudio it will tell you which models will fit in what you've got available.</p>
]]></description><pubDate>Tue, 12 May 2026 12:57:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=48107607</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=48107607</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48107607</guid></item><item><title><![CDATA[New comment by prettyblocks in "Running local models on an M4 with 24GB memory"]]></title><description><![CDATA[
<p>It's great, but I wish I could use these things without it feeling like my laptop is going to melt through the desk.</p>
]]></description><pubDate>Tue, 12 May 2026 12:54:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=48107576</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=48107576</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48107576</guid></item><item><title><![CDATA[New comment by prettyblocks in "The woes of sanitizing SVGs"]]></title><description><![CDATA[
<p>You can also use JS to inject a CSP and it will enforce the strict most policy.</p>
]]></description><pubDate>Tue, 28 Apr 2026 19:52:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=47939698</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=47939698</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47939698</guid></item><item><title><![CDATA[New comment by prettyblocks in "Running local LLMs offline on a ten-hour flight"]]></title><description><![CDATA[
<p>There are paid solutions for this, I've seen Exa recommended.</p>
]]></description><pubDate>Tue, 28 Apr 2026 13:09:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=47934068</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=47934068</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47934068</guid></item><item><title><![CDATA[New comment by prettyblocks in "Running local LLMs offline on a ten-hour flight"]]></title><description><![CDATA[
<p>I've been relatively impressed with local llms, but by far the hardest thing for me is how hot they make my laptop run.</p>
]]></description><pubDate>Tue, 28 Apr 2026 13:08:27 +0000</pubDate><link>https://news.ycombinator.com/item?id=47934054</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=47934054</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47934054</guid></item><item><title><![CDATA[New comment by prettyblocks in "I bought Friendster for $30k – Here's what I'm doing with it"]]></title><description><![CDATA[
<p>I believe it's 9k/year in parking revenue.</p>
]]></description><pubDate>Sun, 26 Apr 2026 23:16:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=47915849</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=47915849</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47915849</guid></item><item><title><![CDATA[New comment by prettyblocks in "Timothy Leary–1960s Acid Guru–May Have Been Among the CIA's Greatest Assets"]]></title><description><![CDATA[
<p>> However, in hindsight, Leary proved to be a false prophet who helped destroy the 1960s movements by pushing young people to take a drug that fried their brains and diverted their energy from political activism.<p>This is a pretty bogus claim. False prophet maybe, but the idea that political activism fizzled out because everyone was eating acid seems pretty unsubstantiated.</p>
]]></description><pubDate>Sat, 25 Apr 2026 17:44:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=47903153</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=47903153</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47903153</guid></item><item><title><![CDATA[New comment by prettyblocks in "Edit store price tags using Flipper Zero"]]></title><description><![CDATA[
<p>Back in the day when I was in HS, kids would go to Borders and swap the stickers from cheap books onto expensive computer books.</p>
]]></description><pubDate>Tue, 21 Apr 2026 14:30:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=47849362</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=47849362</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47849362</guid></item><item><title><![CDATA[New comment by prettyblocks in "I ran Gemma 4 as a local model in Codex CLI"]]></title><description><![CDATA[
<p>For me the vision/OCR is much better than other models in weights class.</p>
]]></description><pubDate>Tue, 14 Apr 2026 13:24:06 +0000</pubDate><link>https://news.ycombinator.com/item?id=47765345</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=47765345</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47765345</guid></item><item><title><![CDATA[New comment by prettyblocks in "Why AI Sucks at Front End"]]></title><description><![CDATA[
<p>> Yes ai can’t see, it only understands numbers.<p>I've also used AI to build frontends that I'm more than satisfied with, and I think it can "see" perfectly fine. The  frontier models are multi-modal and pretty good at vision. You can hook up your coding harness to your browser which will take screenshots of your rendered frontend and modify the code accordingly.</p>
]]></description><pubDate>Mon, 13 Apr 2026 20:42:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=47757550</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=47757550</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47757550</guid></item><item><title><![CDATA[New comment by prettyblocks in "Doom, Played over Curl"]]></title><description><![CDATA[
<p>I think it's safe to assume that "AI help was used" in everything these days and I don't understand why that should be a problem.</p>
]]></description><pubDate>Mon, 13 Apr 2026 20:36:24 +0000</pubDate><link>https://news.ycombinator.com/item?id=47757490</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=47757490</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47757490</guid></item><item><title><![CDATA[New comment by prettyblocks in "I went to America's worst national parks so you don't have to"]]></title><description><![CDATA[
<p>I think it really depends what time of year you go. Zion is beautiful, but when I went it was early in the season and a lot of the popular stuff was closed and it was very crowded everywhere. On the same trip I did all 5 parks in Utah, and I easily rank Zion as my least favorite, even though people seem to love it the most. I would definitely go back though.</p>
]]></description><pubDate>Mon, 13 Apr 2026 15:54:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=47753893</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=47753893</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47753893</guid></item><item><title><![CDATA[New comment by prettyblocks in "Who is Satoshi Nakamoto? My quest to unmask Bitcoin's creator"]]></title><description><![CDATA[
<p>In this case it's not basic stuff. You would need to prove that you own the actual bitcoin or transfer it for it to be collateral on a loan. It's the same as spending it.</p>
]]></description><pubDate>Thu, 09 Apr 2026 20:50:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=47709844</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=47709844</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47709844</guid></item><item><title><![CDATA[New comment by prettyblocks in "Who is Satoshi Nakamoto? My quest to unmask Bitcoin's creator"]]></title><description><![CDATA[
<p>Adam Back is already a high profile target. Unmasking him as Satoshi doesn't really change that for the guy that founded that company that leads bitcoin core development.</p>
]]></description><pubDate>Thu, 09 Apr 2026 20:48:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=47709814</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=47709814</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47709814</guid></item><item><title><![CDATA[New comment by prettyblocks in "Running Gemma 4 locally with LM Studio's new headless CLI and Claude Code"]]></title><description><![CDATA[
<p>how is codex sandbox different from /sandbox on claude code?</p>
]]></description><pubDate>Mon, 06 Apr 2026 01:50:52 +0000</pubDate><link>https://news.ycombinator.com/item?id=47656056</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=47656056</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47656056</guid></item><item><title><![CDATA[New comment by prettyblocks in "A $20/month user costs OpenAI $65 in compute. AI video is a money furnace"]]></title><description><![CDATA[
<p>It's the AWS/Serverless model.</p>
]]></description><pubDate>Fri, 03 Apr 2026 00:50:45 +0000</pubDate><link>https://news.ycombinator.com/item?id=47622119</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=47622119</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47622119</guid></item><item><title><![CDATA[New comment by prettyblocks in "Ask HN: AI productivity gains – do you fire devs or build better products?"]]></title><description><![CDATA[
<p>I think good results come from spending a substantial amount of time reading the claude-code documentation and setting up an environment that lets you work in this new paradigm. Asking it a few things, getting poor results, and deciding "it doesn't work when you ask it to correct itself won't get you that far.</p>
]]></description><pubDate>Mon, 23 Mar 2026 13:39:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=47489397</link><dc:creator>prettyblocks</dc:creator><comments>https://news.ycombinator.com/item?id=47489397</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47489397</guid></item></channel></rss>