<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: Peterz_shu</title><link>https://news.ycombinator.com/user?id=Peterz_shu</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 18 Jun 2026 10:12:49 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=Peterz_shu" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by Peterz_shu in "Ask HN: Do you remember when you gained consciousness? What was it like?"]]></title><description><![CDATA[
<p>I think it was just a chill summer day when I was 5 or sth...<p>Not all conciousness-gaining moments are super climatic</p>
]]></description><pubDate>Wed, 17 Jun 2026 04:50:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=48565828</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=48565828</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48565828</guid></item><item><title><![CDATA[New comment by Peterz_shu in "[dead]"]]></title><description><![CDATA[
<p>I used to think that the best way of getting users is to launch on places like ProductHunt.<p>According to those gurus and ai, "Product Hunt has the exact target user you're looking for". Wrong. People browsing on launch platforms are so crammed with information, that makes it hard for your particular app to gain much users for 2 reasons:<p>1. Not everyone is here to buy. Most people are there to launch their own stuff. So the high-quality people turns into low-quality traffic.<p>2. There are just so many products that only 1-2 products that Product Hunters see will ever be bought by them.<p>That's the cold reality I've realized...<p>So, I'm starting to see it more as a method of gaining credibility. That's why the badges matter so much now.</p>
]]></description><pubDate>Wed, 17 Jun 2026 03:53:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=48565491</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=48565491</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48565491</guid></item><item><title><![CDATA[New comment by Peterz_shu in "Show HN: I built an email agent for founders who are stuck in email"]]></title><description><![CDATA[
<p>Update: Dirac just launched (Woohoo!!)</p>
]]></description><pubDate>Tue, 16 Jun 2026 10:19:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=48553061</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=48553061</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48553061</guid></item><item><title><![CDATA[Show HN: I built an email agent for founders who are stuck in email]]></title><description><![CDATA[
<p>It all started 4 months ago during my last startup. 
Even with barely any traction, I was getting about 20 emails a day. Some from cold randoms, some internal, some from prospects. And I just wasted so much time in my inbox.<p>The current solutions? Superhuman and HEY. Sure they're pretty but the ai is just an after thought. And the tool that was supposed to help me just makes me, my own assistant.<p>So I went and built Dirac.<p>We're launching in 8 hours, 
<a href="https://www.producthunt.com/products/dirac-3?launch=dirac-4" rel="nofollow">https://www.producthunt.com/products/dirac-3?launch=dirac-4</a></p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=48546808">https://news.ycombinator.com/item?id=48546808</a></p>
<p>Points: 3</p>
<p># Comments: 1</p>
]]></description><pubDate>Mon, 15 Jun 2026 20:45:29 +0000</pubDate><link>https://www.dirac.app/</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=48546808</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48546808</guid></item><item><title><![CDATA[New comment by Peterz_shu in "Ask HN: What does your local LLM setup looks like?"]]></title><description><![CDATA[
<p>I have a Mac mini (the square one, idk the model) with M3 chip.<p>Runs pretty well with Ollama on the Qwen models. It seems like Qwen has done a great job with speed.</p>
]]></description><pubDate>Mon, 15 Jun 2026 11:19:06 +0000</pubDate><link>https://news.ycombinator.com/item?id=48539674</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=48539674</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48539674</guid></item><item><title><![CDATA[New comment by Peterz_shu in "Show HN: I used Claude Mythos to build my startup in 1 day"]]></title><description><![CDATA[
<p>This sounds like a traditionally B2B business, wrapped around with a B2C lanidng page blanket.<p>Solid idea, I would lowkey use it.</p>
]]></description><pubDate>Mon, 15 Jun 2026 11:17:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=48539650</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=48539650</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48539650</guid></item><item><title><![CDATA[New comment by Peterz_shu in "Why does paper fold so well?"]]></title><description><![CDATA[
<p>It has a lot of air gaps within its fibers.<p>We know that Solids CANNOT be compressed. So what's actually being folded is the air gaps.<p>Which is why you can't easily fold a piece of tungsten. It has less air gaps.</p>
]]></description><pubDate>Mon, 15 Jun 2026 11:14:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=48539630</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=48539630</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48539630</guid></item><item><title><![CDATA[New comment by Peterz_shu in "Anthropic's Safety Superpower"]]></title><description><![CDATA[
<p>This is the part where the USA and allied countries can gain a headstart from using such an overpowered model.<p>This only just shows how strong Mythos/Fable will be, once released to the public.<p>I'm guessing about 0.5 year till public.</p>
]]></description><pubDate>Mon, 15 Jun 2026 11:13:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=48539614</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=48539614</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48539614</guid></item><item><title><![CDATA[New comment by Peterz_shu in "Ask HN: What are you working on? (June 2026)"]]></title><description><![CDATA[
<p>100% of the world's current email apps are made for users to spend time within it.<p>But the "ai email app" that should've saved time for them, ends up making the user their own assistant.<p>So I build <a href="https://dirac.app" rel="nofollow">https://dirac.app</a> , a decision-based, ai-native email app.<p>This will be the Next Superhuman, and it's launching tomorrow on ProductHunt.</p>
]]></description><pubDate>Mon, 15 Jun 2026 11:11:30 +0000</pubDate><link>https://news.ycombinator.com/item?id=48539594</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=48539594</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48539594</guid></item><item><title><![CDATA[New comment by Peterz_shu in "Ask HN: How do you manage your morning catch-up routine?"]]></title><description><![CDATA[
<p>True. But how do I know if it will be caught later anyways? I'd just end up getting distracted in the middle of my work, which is arguably even more distracting, right?</p>
]]></description><pubDate>Wed, 21 Jan 2026 09:39:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=46703230</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=46703230</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46703230</guid></item><item><title><![CDATA[New comment by Peterz_shu in "Ask HN: How do you manage your morning catch-up routine?"]]></title><description><![CDATA[
<p>Yeah I haven't looked into dashboarding, as i always thought it took too long and was too limited. But i will try now thanks :)</p>
]]></description><pubDate>Wed, 21 Jan 2026 09:38:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=46703222</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=46703222</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46703222</guid></item><item><title><![CDATA[Ask HN: How do you manage your morning catch-up routine?]]></title><description><![CDATA[
<p>Every day I spend 20-40 minutes just to check dozens of apps. Github commits, Discord community, insta cofounder messages, Stripe payments, and so on...<p>It's not hard work, but it's friction before the real work starts.<p>Curious how others handle this.
Do you have a system? An app? Or just accept the tax :(</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46702458">https://news.ycombinator.com/item?id=46702458</a></p>
<p>Points: 2</p>
<p># Comments: 4</p>
]]></description><pubDate>Wed, 21 Jan 2026 07:53:42 +0000</pubDate><link>https://news.ycombinator.com/item?id=46702458</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=46702458</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46702458</guid></item><item><title><![CDATA[New comment by Peterz_shu in "Show HN: I treated dating like a Seed Round. Here is the Term Sheet"]]></title><description><![CDATA[
<p>If dating is a seed round, then you are your product. And product is the most important in a startup.<p>So, how to make product good?  :)</p>
]]></description><pubDate>Mon, 19 Jan 2026 04:27:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=46675062</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=46675062</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46675062</guid></item><item><title><![CDATA[New comment by Peterz_shu in "Anyone still planning their holidays in Sheets? Try this instead"]]></title><description><![CDATA[
<p>Great app idea, and here are three points for improvement:<p>- We don't need that many steps in the how-to-use popups. Just the 4 steps in the beginning is plenty (more is overwhelming)<p>- Maybe add some images related to my trip and make the site look less "daunting"<p>- Add an ai chatbox so I can ask an ai agent for it to directly edit my trip plans<p>Hope this helps :)</p>
]]></description><pubDate>Mon, 19 Jan 2026 04:25:08 +0000</pubDate><link>https://news.ycombinator.com/item?id=46675050</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=46675050</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46675050</guid></item><item><title><![CDATA[New comment by Peterz_shu in "Claude voice mode is still a joke in 2026"]]></title><description><![CDATA[
<p>This is what I really don't get: What's wrong with just typing? Its more private, and can be efficiently documented and shared (unlike a voice agent, which has to waste energy translating audio back to text).<p>I mean... Wispr flow is still great tho</p>
]]></description><pubDate>Mon, 19 Jan 2026 04:19:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=46675025</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=46675025</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46675025</guid></item><item><title><![CDATA[New comment by Peterz_shu in "Show HN: PixelRipple – AI ads agent for e-commerce"]]></title><description><![CDATA[
<p>I noticed that too; Do you mean in the "Real results, Real revenue" section on the website?</p>
]]></description><pubDate>Mon, 19 Jan 2026 04:18:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=46675017</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=46675017</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46675017</guid></item><item><title><![CDATA[New comment by Peterz_shu in "Show HN: PixelRipple – AI ads agent for e-commerce"]]></title><description><![CDATA[
<p>Sounds great if it is dones right (I will start testing it soon)<p>What does the ai generate as in "creatives"? Is it ai-generated media, or just ai-directed?</p>
]]></description><pubDate>Mon, 19 Jan 2026 04:17:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=46675007</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=46675007</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46675007</guid></item><item><title><![CDATA[New comment by Peterz_shu in "[dead]"]]></title><description><![CDATA[
<p>I'm a high school student. My co-founder and I work on our startup for ~1 hour per day between homework and sleep.<p>The problem we kept hitting: Sitting down to code at 9pm, spending 15 minutes checking:
- GitHub (what got pushed?)
- Gmail (any user issues?)
- Discord (team updates?)
- Analytics (signups, errors, metrics)
- Trying to remember what I was working on<p>Then finally starting to code at 9:15pm with 45 minutes left.<p>So we built Dirac. One command pulls from your connected services and gives you a plain-text summary of what happened since you last checked.<p>*Tech details:*
- local-hosted widget
- OAuth integrations with GitHub, Gmail, Discord
- Local caching for speed
- Written in python + electron
- Currently pulls: commits, PR comments, emails matching filters, team messages, basic metrics<p>*Current limitations (being honest):*
- Only works for solo/small teams (not enterprise-scale)
- Summary quality varies (working on better filtering)
- Setup takes ~5 minutes (too long)
- No mobile app yet<p>*Why we're launching at 80%:*
This is our third pivot. First tried building a general AI agent (too hard). Then task automation (no market). Now solving our own problem. We needed to validate before spending 3 more months building.<p>Launched on Product Hunt today (currently #18). Learning that launching is more about conversation than ranking.<p>For HN: What am I missing technically? What would make this actually useful vs. just another aggregator?<p>Demo: <a href="https://www.youtube.com/watch?v=Fa5ddZ1achg" rel="nofollow">https://www.youtube.com/watch?v=Fa5ddZ1achg</a><p>Open to technical critique—that's why we're here.</p>
]]></description><pubDate>Tue, 06 Jan 2026 22:07:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=46519531</link><dc:creator>Peterz_shu</dc:creator><comments>https://news.ycombinator.com/item?id=46519531</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46519531</guid></item></channel></rss>