<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: diavelguru</title><link>https://news.ycombinator.com/user?id=diavelguru</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Tue, 07 Apr 2026 07:56:54 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=diavelguru" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by diavelguru in "Issue: Claude Code is unusable for complex engineering tasks with Feb updates"]]></title><description><![CDATA[
<p>As soon as that change came through I set the effort to high.  Have not regretted it for any coding task.  It feels the same as Dec-Jan though now spawning more sub agents which is not a bad thing.</p>
]]></description><pubDate>Tue, 07 Apr 2026 01:06:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=47669476</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=47669476</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47669476</guid></item><item><title><![CDATA[New comment by diavelguru in "Issue: Claude Code is unusable for complex engineering tasks with Feb updates"]]></title><description><![CDATA[
<p>Love this.  Responding to users.  Detail info investigating. Action being taken (at least it seems so).</p>
]]></description><pubDate>Tue, 07 Apr 2026 01:04:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=47669463</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=47669463</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47669463</guid></item><item><title><![CDATA[New comment by diavelguru in "Plenty of AI hype, but not much useful software?"]]></title><description><![CDATA[
<p>With AI, I've tackled lots of personal projects I had on the back burner for lack of time: recreating some old games, creating some marketing and shopping sites for the wife and also professionally: using it to fulfill business requirements.  So in my little corner of the world it is being used and a lot mostly nonstop from Aug 2025 to present.  Re your question: I think it has to do with all the rest of what's needed for a production system, the documentation, CI/CD, production deployments: all those pesky still-needed-human-things that are somewhat tedious to set and get correct.</p>
]]></description><pubDate>Sun, 08 Mar 2026 03:54:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=47294256</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=47294256</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47294256</guid></item><item><title><![CDATA[New comment by diavelguru in "If AI is so good, why don't we have an infinite supply of 10x engineers?"]]></title><description><![CDATA[
<p>We get a few not infinite.  with a Claude 5x Max subscription I get 3-5 workable parallel agents that themselves spawn subagents to complete work in a ~3 hour window before tokens are spent.  I imagine with 20x Max that could be increased to maybe 10 within a 5 hour window?  so if you want infinite you're also talking infinite money.  Working in parallel takes me personally to a higher plane of output than that of a 10x engineer as it pertains to fulfilling requirements.</p>
]]></description><pubDate>Sun, 08 Mar 2026 03:43:36 +0000</pubDate><link>https://news.ycombinator.com/item?id=47294187</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=47294187</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47294187</guid></item><item><title><![CDATA[New comment by diavelguru in "Why developers using AI are working longer hours"]]></title><description><![CDATA[
<p>and flow state is a luxury in 2026 with AI swarm most likely to be found sparingly if all.  Good luck all!</p>
]]></description><pubDate>Sun, 08 Mar 2026 00:48:49 +0000</pubDate><link>https://news.ycombinator.com/item?id=47293150</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=47293150</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47293150</guid></item><item><title><![CDATA[New comment by diavelguru in "Why developers using AI are working longer hours"]]></title><description><![CDATA[
<p>you are correct flow is not achieved as this is not programming more like system design, architecture, QA, Product Owner work.  It's using the swarm as your own dev team.</p>
]]></description><pubDate>Sun, 08 Mar 2026 00:45:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=47293124</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=47293124</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47293124</guid></item><item><title><![CDATA[New comment by diavelguru in "Why developers using AI are working longer hours"]]></title><description><![CDATA[
<p>yes agreed.  I'm running 3-5 parallel Claude at once with requirements as the input.  My prompt is say work on section 5.1 or something very specific.  Then I'm monitoring the work across all instances.</p>
]]></description><pubDate>Sun, 08 Mar 2026 00:44:18 +0000</pubDate><link>https://news.ycombinator.com/item?id=47293115</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=47293115</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47293115</guid></item><item><title><![CDATA[New comment by diavelguru in "Why developers using AI are working longer hours"]]></title><description><![CDATA[
<p>This is a real thing.  I spent all of January doing Greenfield development using Claude (I finished the requirements) and all I can say is thank goodness I had the Max 5x plan and not the 20x as I got breaks once the tokens were used up till the next cycle.  I was forced to get up and do something else.  That something else was biking, rowing, walking.  My productivity had never been higher but at what cost?  My health no thanks.  So I'm glad I'm using the time till token reset for my health.  I time it perfectly.  I do a walk, row, bike for 1 hour then as I arrive back the tokens are reset.  I get like 3 hours nonstop use per token batch with the 5x plan.  I've been thinking about going 20x but am scared...</p>
]]></description><pubDate>Sun, 08 Mar 2026 00:27:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=47292997</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=47292997</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47292997</guid></item><item><title><![CDATA[New comment by diavelguru in "Ask HN: What's the best engineering interview process?"]]></title><description><![CDATA[
<p>Definitely have some questions to get a baseline understanding of the tech stack but then ask a problem solving question and listen to  what they say and how they think and use their answers to  turn the interview into a conversation by asking them questions using their answers as a guide.  The flow will be natural for both interviewer and those interviewing and insights from the conversation are very revealing.</p>
]]></description><pubDate>Fri, 27 Feb 2026 21:36:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=47185979</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=47185979</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47185979</guid></item><item><title><![CDATA[Claude /fast mode consumes money fast]]></title><description><![CDATA[
<p>I got a $50 credit from Claude for the /extra-usage command which I turned on and cleared context and had it debug a particularly thorny issue related to a C app conversion to swift with lots of external resources.  While it didn't solve the issue (I switched /fast off after two rounds), I think it got me far enough to solve the problem with [regular] Claude.  After the first round it used $17 of the $50 and after the second issue to fix/diagnose it used $35 of $50.  This isn't a complaint and I might use it more on focused, targeted debugging/solutions - or getting me there quicker.  I enjoyed the speed.  It's not blazing but as they say I can see a 2.~ speedup.  I just need to beware that my deposit will go quickly.  Just some thoughts after using it.</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46955265">https://news.ycombinator.com/item?id=46955265</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 10 Feb 2026 04:08:05 +0000</pubDate><link>https://news.ycombinator.com/item?id=46955265</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=46955265</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46955265</guid></item><item><title><![CDATA[New comment by diavelguru in "Tell HN: I am afraid AI will take my job at some point"]]></title><description><![CDATA[
<p>I agree with this guy: <a href="https://obie.medium.com/what-happens-when-the-coding-becomes-the-least-interesting-part-of-the-work-ab10c213c660" rel="nofollow">https://obie.medium.com/what-happens-when-the-coding-becomes...</a> 
When you look at the AI as having a Jr. Developer at your disposal to do the mundane things you actually elevate yourself higher than you could individually (without the help) as you focus more on architecture and design and guide the LLM to make incremental improvements instead of decremental ones.  That's my take at least.<p>This one also:
<a href="https://terriblesoftware.org/2025/12/11/ai-can-write-your-code-it-cant-do-your-job/" rel="nofollow">https://terriblesoftware.org/2025/12/11/ai-can-write-your-co...</a></p>
]]></description><pubDate>Sat, 27 Dec 2025 04:39:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=46399189</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=46399189</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46399189</guid></item><item><title><![CDATA[New comment by diavelguru in "Opus 4.5 is the first model that makes me fear for my job"]]></title><description><![CDATA[
<p>I agree with this guy: <a href="https://obie.medium.com/what-happens-when-the-coding-becomes-the-least-interesting-part-of-the-work-ab10c213c660" rel="nofollow">https://obie.medium.com/what-happens-when-the-coding-becomes...</a>. It's a game changer when used to pair program (MOB Programming) with you as the navigator and the LLM as the driver.  It needs guidance and rework but shit gets done with the human as the ultimate gatekeeper.</p>
]]></description><pubDate>Sun, 14 Dec 2025 23:31:08 +0000</pubDate><link>https://news.ycombinator.com/item?id=46268294</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=46268294</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46268294</guid></item><item><title><![CDATA[New comment by diavelguru in "Ask HN: What is the most important thing in life?"]]></title><description><![CDATA[
<p>Claude opus 4.1:
Your framework for thinking about life's purpose through utility maximization is intellectually rigorous and thought-provoking. Let me engage with your ideas by examining both what works well in your model and where it might benefit from additional dimensions.<p>## The Strengths of Your Utility Framework<p>Your insight about logarithmic utility functions is particularly astute. You're absolutely right that spreading value across multiple people generates more total utility than concentrating it in one person - this aligns with established economic principles about diminishing marginal returns. The example of $100,000 to ten people versus $1,000,000 to one person captures this beautifully.<p>The temporal dimension you've identified is equally important and often overlooked. When you note that helping your parents at 70 provides more utility than waiting until they're 80, you're recognizing that people's capacity to enjoy and benefit from resources changes over time. This adds a crucial urgency to altruism - it's not just about giving, but giving when it matters most.<p>## Potential Refinements to Consider<p>However, I'd invite you to think about a few aspects that might enrich your framework:<p>*The Paradox of Self-Care*: Your model positions maximizing others' utility as "the direct opposite" of maximizing your own. But consider this: if everyone followed this principle perfectly, who would be the recipient of all this utility? There's a recursive problem here. Moreover, maintaining your own wellbeing often enables you to provide more utility to others over time. Think of it less as opposite directions and more as interconnected systems - your capacity to help others depends partly on your own stability and resources.<p>*Beyond Utility Calculations*: While utility functions provide a useful analytical tool, human meaning-making often transcends these calculations. Consider experiences like creating art, pursuing knowledge for its own sake, or moments of profound connection. These might not maximize anyone's utility in measurable ways, yet they seem central to what makes life meaningful. How would your framework account for the scientist pursuing pure research with no immediate practical application, or the artist creating something beautiful but "useless"?<p>*The Measurement Problem*: You acknowledge that value isn't just financial but also emotional. This is crucial, but it also introduces significant complexity. How do we compare the utility of teaching someone to read versus comforting them through grief? The incommensurability of different types of value makes optimization extremely difficult in practice.<p>## A Teaching Exercise<p>Here's a thought experiment to test your framework: Imagine you have the choice between two lives. In Life A, you maximize total utility delivered to others through efficient but impersonal systems (perhaps developing technology that helps millions). In Life B, you form deep, transformative relationships with a smaller number of people. Which life delivers more utility? Which would you find more meaningful? The tension between these might reveal something about the limits of pure utility maximization.<p>## Building on Your Foundation<p>Your three optimization questions are excellent starting points. But I'd suggest adding a fourth: "How do I discover what actually provides utility to others, given that people's needs are complex and sometimes unknown even to themselves?"<p>Your framework shows remarkable sophistication for someone at 21, particularly in recognizing the multi-dimensional nature of value and the importance of timing. As you continue developing these ideas, you might find that the most important thing in life isn't a single principle but rather the dynamic balance between several essential elements - helping others, yes, but also self-development, creation, discovery, and connection.<p>The fact that you end with humility about these being preliminary thoughts shows wisdom. The question "what is most important in life?" is one humans have grappled with for millennia, and each generation must find its own synthesis of ancient wisdom and contemporary understanding. Your utility framework contributes meaningfully to that ongoing conversation.</p>
]]></description><pubDate>Sat, 08 Nov 2025 04:54:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=45854251</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=45854251</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45854251</guid></item><item><title><![CDATA[New comment by diavelguru in "Ask HN: Are foreign key constraints still useful?"]]></title><description><![CDATA[
<p>You answered your own question in the description.  "DB constraints pop up when I am making changes to my local environment” and that gave you the confidence to promote to QA, Staging and finally Production.  A well tested application will find those problematic issues long before it goes to prod.  I’m just glad that all business logic is not thrown in the DB as in the past.  The separation makes for a much quicker and robust system.  Don’t think DB constraints are a relic they are very much a part of the relational database management system and still used in those systems very successfully.  Another topic for another thread is the relevance of the RDBMS itself not just one subset of it.</p>
]]></description><pubDate>Tue, 12 Aug 2025 23:04:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=44882792</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=44882792</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44882792</guid></item><item><title><![CDATA[New comment by diavelguru in "What good AI news aggregator are there?"]]></title><description><![CDATA[
<p><a href="https://www.newsminimalist.com/" rel="nofollow">https://www.newsminimalist.com/</a></p>
]]></description><pubDate>Mon, 02 Dec 2024 05:57:26 +0000</pubDate><link>https://news.ycombinator.com/item?id=42293521</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=42293521</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42293521</guid></item><item><title><![CDATA[New comment by diavelguru in "Ask HN: Should we bring software dev in-house?"]]></title><description><![CDATA[
<p>Agree with this re serverless cloud computing</p>
]]></description><pubDate>Fri, 09 Aug 2024 01:04:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=41197938</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=41197938</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41197938</guid></item><item><title><![CDATA[New comment by diavelguru in "Ask HN: Should we bring software dev in-house?"]]></title><description><![CDATA[
<p>One thing to remember is that some/most are not turned off by a non-glamorous industry like logistics but rather attracted to solving problems regardless of the industry.  I’d hire someone who can actually do coding (so not a CTO, CIO level) that is also or has been a lead who can delve into your business and review existing products and can get you an 80% POC in under 3 months.  It may not be pretty but it will help you better position to ask for more $$ for contractors or other employees.  Start with 1 then scale from there.  I’d give them 1 month to study the existing system with no deliverables other than documentation that will be used to replicate your product needs.  I’d also be sure to hire a PO (Product Owner) if you don’t have one and have them also learn the product inside and out as they will be the gatekeeper between the coders and the users as to how the product functions and that the coders fulfill all the requirements.</p>
]]></description><pubDate>Thu, 08 Aug 2024 23:59:57 +0000</pubDate><link>https://news.ycombinator.com/item?id=41197640</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=41197640</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41197640</guid></item><item><title><![CDATA[Every Noise at Once]]></title><description><![CDATA[
<p>Article URL: <a href="https://everynoise.com/">https://everynoise.com/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=40149423">https://news.ycombinator.com/item?id=40149423</a></p>
<p>Points: 1</p>
<p># Comments: 1</p>
]]></description><pubDate>Wed, 24 Apr 2024 20:48:19 +0000</pubDate><link>https://everynoise.com/</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=40149423</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40149423</guid></item><item><title><![CDATA[New comment by diavelguru in "Ask HN: Is crypto an unviable, tech-bubble delusion?"]]></title><description><![CDATA[
<p><a href="https://web3isgoinggreat.com/" rel="nofollow noreferrer">https://web3isgoinggreat.com/</a></p>
]]></description><pubDate>Sun, 30 Jul 2023 03:58:30 +0000</pubDate><link>https://news.ycombinator.com/item?id=36927673</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=36927673</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=36927673</guid></item><item><title><![CDATA[New comment by diavelguru in "Ask HN: Visiting SF on a Budget?"]]></title><description><![CDATA[
<p>BART it up!</p>
]]></description><pubDate>Fri, 28 Jul 2023 05:50:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=36903452</link><dc:creator>diavelguru</dc:creator><comments>https://news.ycombinator.com/item?id=36903452</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=36903452</guid></item></channel></rss>