<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: tinyhouse</title><link>https://news.ycombinator.com/user?id=tinyhouse</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Wed, 18 Nov 2026 04:07:29 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=tinyhouse" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by tinyhouse in ""Who is Hiring?" postings stuck near 12-year lows as AI reaches half of all jobs"]]></title><description><![CDATA[
<p>Developers who get laid off today have very hard time finding something new. The bar is rising and if you don't stand out you're not getting a job. In the US if you're outside NYC and SF it's really bad. If you need a job make don't get discouraged if it takes you more than six months finding someone new. It's the norm now if you don't have a unicorn on your resume.</p>
]]></description><pubDate>Mon, 16 Nov 2026 03:10:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=50513581</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=50513581</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50513581</guid></item><item><title><![CDATA[New comment by tinyhouse in "Val Town is winding down"]]></title><description><![CDATA[
<p>I actually created one simple agent back then on Val Town that checks any new sign ups to my app and then posts to Slack. It was free but I've never created anything else on their platform.<p>I kind of liked their story and wanted to support them. Their UX wasn't great for one (back when I tried it at least). But I think the bigger problem is that the big opportunity for them was that they started early before it was crowded, and if they didn't capture a big market share then, now it's too late and impossible to compete. For agents to be useful you need to integrate them with all your services and data, and if you're already using a trusted provider Like Claude or CloudFlare etc. then it's easy just to stick with their offerings.<p>Thanks for the migration guide!</p>
]]></description><pubDate>Tue, 10 Nov 2026 21:38:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=50446672</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=50446672</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50446672</guid></item><item><title><![CDATA[New comment by tinyhouse in "Perplexity launches AI agents for work"]]></title><description><![CDATA[
<p>They have users and revenue in the hundreds of millions so yes. No affiliation. I don't use their service either.<p>I also think that this is going to be very useful and many startups are entering this space of agent teammates with persistent memory.</p>
]]></description><pubDate>Fri, 06 Nov 2026 18:53:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=50396646</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=50396646</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50396646</guid></item><item><title><![CDATA[New comment by tinyhouse in "Nasdaq Sinks 5% as Investors Retreat from AI Trade After Election"]]></title><description><![CDATA[
<p>Great opportunity to buy the dip in a couple of weeks. If anyone thinks they can slow down or stop AI adoption by banning data centers, good luck with that... The Chinese built an AI empire with much less compute that exists today in the US.<p>Also, people don't have issue with data centers. They have issues with AI cause they think it's gonna make them unemployed, which is a real worry I can relate to. I just don't think that banning data centers or AI progress in general would help anyone keeping their job.</p>
]]></description><pubDate>Thu, 05 Nov 2026 19:21:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=50383218</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=50383218</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50383218</guid></item><item><title><![CDATA[New comment by tinyhouse in "Ask HN: What's your current AI agentic setup?"]]></title><description><![CDATA[
<p>For personal work. I use Orca.<p>1. Claude Code with $20 Claude subscription. This used to be a terrible value but now it's a great value, probably because of the IPO. I do a lot of coding and other things and I rarely get close to my weekly limit. I mostly use Opus 5.5 medium. Codex with gpt used to be good value but recently I canceled it cause I kept reaching my limits.<p>2. When I want a "different pair of eyes" to review things I'm working on with Opus, I use Pi with other model providers and just pay API prices.<p>3. Deepseek harness with a deepseek API key. Really great value.<p>4. I have a Muse $5/month subscription that I keep as a backup although I don't really need it.<p>I also use cloud agents to setup reoccurring jobs. For expensive agent runs, like visit 5000 websites and extract something from each, I would set up something cheap with an agents sdk and for the model use either Muse, Deepseek, or GPT-6-luna using their batch API.<p>I maintain all my dotfiles and agent files / context in a github repo so that I can switch easily between harnesses and providers.</p>
]]></description><pubDate>Wed, 04 Nov 2026 16:16:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=50367744</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=50367744</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50367744</guid></item><item><title><![CDATA[New comment by tinyhouse in "US tech stocks see sharp falls over AI cost fears"]]></title><description><![CDATA[
<p>I think the assumption that AI demand will keep growing is a no-brainer based on the value that AI brings. However, you need people and companies to pay for that demand, and if people and companies start losing money due to AI and job cuts, then you have demand but not enough money to pay for it.</p>
]]></description><pubDate>Fri, 30 Oct 2026 17:56:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=50307693</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=50307693</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50307693</guid></item><item><title><![CDATA[New comment by tinyhouse in "Claude Duet is the first personal-assistant-type harness that feels useful"]]></title><description><![CDATA[
<p>Sounds like Duet is more like Grokbot than Muse? Haven't used it but the author is talking about channels, github policy access for specific channels, etc. That doesn't sound like the Muse use case.<p>I actually use Claude's agents on the Claude platform for all things work related and it works pretty well. It's not cheap though if you use the more expensive models and setting it up requires currently some initial effort for connecting integrations like Slack the average user is unlikely to do. It also doesn't let you change things. Every change requires a new deployment. As a cloud agent that runs on a schedule it's pretty good. I think they just need to figure out a single place for this. Currently you can set up agents on the claude platform, as part of claude code, and as part of regular claude.</p>
]]></description><pubDate>Thu, 29 Oct 2026 21:44:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=50296542</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=50296542</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50296542</guid></item><item><title><![CDATA[New comment by tinyhouse in "Microsoft offers mass buyouts for roughly a quarter of its workforce"]]></title><description><![CDATA[
<p>The writing was on the wall for a long time. All of big tech (and many midsize tech too) have way too many employees. In the AI era is just more absurd than it was before. I always appreciate companies that take care of their employees this way so kudos to Microsoft. We haven't seen the package they are offering yet but I'm assuming it'd going to be very good, esp for those who have been there for a while.</p>
]]></description><pubDate>Mon, 26 Oct 2026 17:02:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=50247106</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=50247106</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50247106</guid></item><item><title><![CDATA[New comment by tinyhouse in "Anthropic S-1"]]></title><description><![CDATA[
<p>Good for you if you have Anthropic shares! I have no clue how they keep their employees around once the 6 month lockup period ends. They all going to be multi-millionaires, either retire or leave to start something on their own.</p>
]]></description><pubDate>Mon, 26 Oct 2026 13:20:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=50243166</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=50243166</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50243166</guid></item><item><title><![CDATA[New comment by tinyhouse in "Roku developer repositories moved from GitHub to GitLab"]]></title><description><![CDATA[
<p>No horse in this race but GitLab has been down a few times recently too.</p>
]]></description><pubDate>Sat, 24 Oct 2026 16:53:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=50223031</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=50223031</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50223031</guid></item><item><title><![CDATA[New comment by tinyhouse in "Arc Is Sunsetting"]]></title><description><![CDATA[
<p>Is this just the product Arc or the entire company? (The Browser Company)</p>
]]></description><pubDate>Thu, 22 Oct 2026 16:51:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=50197939</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=50197939</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50197939</guid></item><item><title><![CDATA[New comment by tinyhouse in "Claude Duet"]]></title><description><![CDATA[
<p>Next week Google is dropping one too.</p>
]]></description><pubDate>Wed, 21 Oct 2026 19:28:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=50186247</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=50186247</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50186247</guid></item><item><title><![CDATA[New comment by tinyhouse in "DeepSeek just open-sourced its inference engine"]]></title><description><![CDATA[
<p>Currently this is true only for Anthropic's $20/month plan. I do a lot of work without reaching my limit. OAI's similar plan on the other hand isn't worth it - it used to be a great plan in the begining when they needed to catch up, but now I reach my limit very quickly on it. The only reason I keep it is because my entire family is using this account for their chatgpt usage.<p>The more expensive subscriptions are just not worth it for individuals. Using deepseek even at API prices is better value. If you're a company it might be a different story.</p>
]]></description><pubDate>Tue, 20 Oct 2026 12:02:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=50165005</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=50165005</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50165005</guid></item><item><title><![CDATA[New comment by tinyhouse in "How employment is changing at Anthropic"]]></title><description><![CDATA[
<p>I'm surprised this is not much more talked about these days. The hiring bar is going to increase so much in the next few years (already happening), and together with more layoffs driven by big tech, things aren't looking good.<p>Anyhropic is not a good example for anything. Their scale and growth is way beyond what like 99% of companies experiece. Their optimization formula is different.</p>
]]></description><pubDate>Sun, 18 Oct 2026 11:51:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=50140246</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=50140246</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50140246</guid></item><item><title><![CDATA[New comment by tinyhouse in "DeepSeek search and fetch – $0.75 per 1000 successful requests"]]></title><description><![CDATA[
<p>DeepSeek models are amazing and this search pricing is fantastic. All these companies like Parallel, Exa and others are gonna have hard time justifying their hefty valuations. Add Perplexity and many others into the mix. The demand for what they are building is there, but not the margins for the amount of money they raised.</p>
]]></description><pubDate>Sun, 18 Oct 2026 11:32:49 +0000</pubDate><link>https://news.ycombinator.com/item?id=50140112</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=50140112</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50140112</guid></item><item><title><![CDATA[New comment by tinyhouse in "DeepSeek-v4.1-Flash-1017"]]></title><description><![CDATA[
<p>Web search: $0.75 per 1,000 calls<p>That's way cheaper than openai and other web search providers. I have no doubt that AI demand justifies all the compute money big tech is spending now, but I also think there are so many over valued VC funded companies that are going to crash in the next few years.</p>
]]></description><pubDate>Sat, 17 Oct 2026 15:02:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=50131751</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=50131751</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50131751</guid></item><item><title><![CDATA[New comment by tinyhouse in "Gemini 4 Argon"]]></title><description><![CDATA[
<p>Google didn't release an impressive model since Flash 3. I tried Antigravity with the latest Gemini model for a week and it's the worse experience I had among more than 5 harnesses and models I tried (deeseek/pro, muse/spark-1.3, cc/opus, codex/sol, and Pi/sol). I haven't touched it since and canceled my pro subscription.<p>Then I tried Flash 3.8 for different tasks like OCR and others, and while in their benchmarks it crashes Flash 3, in my experiments I didn't notice much difference, often even Flash 3 performed better.<p>I hope Gemini 4 Argon is a real step up from that but we'll see once they release it. I'm rooting for Google and it's about time they deliver frontier intelligence, not just competitive prices.</p>
]]></description><pubDate>Thu, 01 Oct 2026 01:13:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=49916498</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=49916498</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49916498</guid></item><item><title><![CDATA[New comment by tinyhouse in "Dots: Always-on agents"]]></title><description><![CDATA[
<p>To Anthropic's credit, I find their overall design so much better than OpenAI's.</p>
]]></description><pubDate>Tue, 29 Sep 2026 17:28:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=49897003</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=49897003</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49897003</guid></item><item><title><![CDATA[New comment by tinyhouse in "Dots: Always-on agents"]]></title><description><![CDATA[
<p>One way that I find the AI space boring is that everyone is working on the same things. The release of Jev was a breath of fresh air.</p>
]]></description><pubDate>Tue, 29 Sep 2026 17:22:36 +0000</pubDate><link>https://news.ycombinator.com/item?id=49896895</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=49896895</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49896895</guid></item><item><title><![CDATA[New comment by tinyhouse in "I built non-autoregressive decision models with RL a year ago"]]></title><description><![CDATA[
<p>You're right but it's not the full picture. It's much easier to market when you have a name brand behind you. Not sure the author would've done much better even if he messaged it better. It's like the difference between someone random saying something smart on Twitter and no one gives a shit and Karapthy saying the same thing and everyone talks about it. I'm not saying it in a bad way - those with clout around them earned the people's trust by doing something right. But it's not easy to get there and there are many people doing great things that get very little publicity if at all. Not to mention in this case Jev came from a startup that raised a lot of money and can spend it on good marketing.</p>
]]></description><pubDate>Sat, 19 Sep 2026 19:37:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=49769513</link><dc:creator>tinyhouse</dc:creator><comments>https://news.ycombinator.com/item?id=49769513</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49769513</guid></item></channel></rss>