<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: js98</title><link>https://news.ycombinator.com/user?id=js98</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Mon, 01 Jun 2026 21:41:19 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=js98" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by js98 in "I made my phone slow on purpose"]]></title><description><![CDATA[
<p>Personally, my recent and surprisingly greatest win was to set up my old phone (samsung S21) with the addictive apps, removing them completely from my iPhone.<p>Quite literally "cold-turkey'ed" from 4.5-ish hours/day to 2 hours a day in a single day, consistent over the last few weeks.<p>I set up my second phone with a custom homescreen, and installing the 'bad' apps on there (Instagram, Youtube, NYTimes in particular). I dont use it for other apps.<p>Now if I want to scroll, which I still do sometimes, I have to walk to a specific chair next to which my 'addiction phone' is, I'll scroll for 10-15 minutes, and get back to the real world. I used to have particular issues with scrolling during vibe-coding sessions, and I'm genuinely surprised how well this approach worked for me.</p>
]]></description><pubDate>Mon, 01 Jun 2026 16:54:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=48359421</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=48359421</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48359421</guid></item><item><title><![CDATA[Ask HN: How much AI is in your writing?]]></title><description><![CDATA[
<p>Looking at tech blogs, HN submissions, et cetera, a lot of content is (partly, increasingly?) generated by AI. The value of the reader's time and slop has been discussed at length. I ask:<p>Among you, the HN-community who writes, how much do you use AI in the process?<p>There is somewhat a spectrum from none, to grammarly-style cleanup, to whisprflow to 'look at my project, write me a HN-ready post'.<p>Personally, I've done a 180, where I started at light touchups to rough draft-to-polished back to almost none, to keep the writing my own (I do some writing at jakobs.dev).<p>As a reader, do you believe you notice, and care?</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=48235223">https://news.ycombinator.com/item?id=48235223</a></p>
<p>Points: 10</p>
<p># Comments: 37</p>
]]></description><pubDate>Fri, 22 May 2026 12:55:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=48235223</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=48235223</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48235223</guid></item><item><title><![CDATA[New comment by js98 in "Ask HN: What Are You Working On? (April 2026)"]]></title><description><![CDATA[
<p>Hey all, I’ve built Nekode (<a href="https://github.com/Jakob-98/nekode" rel="nofollow">https://github.com/Jakob-98/nekode</a>), which is desktop-cats for monitoring your AI coding agent sessions (OpenCode/Claude Code/ Copilot CLI/VSCode Copilot). It works as a macOS menubar app, and the cats run around on your desktop.<p>Each cat mirrors the agent's state, such as sleeping when idle, walking when working, sitting when waiting for input, running toward your cursor when it needs permission.<p>Fully native Swift, no Electron, under 5 MB, zero network requests, all session data stays local as plain JSON.<p>I published it source-available with an honor-system license, but this week I’m going to fully open source it and remove the licence. The payment/nag system was an interesting experiment but the project is more useful to me as a proper OSS tool at this point.<p><a href="https://nekode.dev" rel="nofollow">https://nekode.dev</a></p>
]]></description><pubDate>Mon, 13 Apr 2026 07:20:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=47748792</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=47748792</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47748792</guid></item><item><title><![CDATA[New comment by js98 in "Obsidian Sync now has a headless client"]]></title><description><![CDATA[
<p>Oh wow thats good timing. I DIY’ed a workaround for this only last week for my openclaw instance. Happy to change it to this!
<a href="https://jakobs.dev/granular-notes-access-limiting-openclaw-blast-radius/" rel="nofollow">https://jakobs.dev/granular-notes-access-limiting-openclaw-b...</a></p>
]]></description><pubDate>Sat, 28 Feb 2026 21:05:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=47200225</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=47200225</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47200225</guid></item><item><title><![CDATA[New comment by js98 in "Pragmatic Notes on Running Dangerous AI Coding Agents in Cloud VMs"]]></title><description><![CDATA[
<p>Hey all. TL;DR: none of this is super novel or fancy, but the set of a few simple approaches (cloud VM, tailscale for connecting, strong isolation, long-running sessions, basic notifications) made using agents with all tools enabled much more useful for me.<p>I hadn’t seen many practical writeups on running coding agents in cloud VMs specifically, so I figured it was worth sharing what actually worked day-to-day for me. I use this approach practically daily now</p>
]]></description><pubDate>Wed, 21 Jan 2026 12:32:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=46704772</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=46704772</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46704772</guid></item><item><title><![CDATA[Pragmatic Notes on Running Dangerous AI Coding Agents in Cloud VMs]]></title><description><![CDATA[
<p>Article URL: <a href="https://jakobs.dev/pragmatic-notes-running-dangerous-ai-agents-cloud-vms/">https://jakobs.dev/pragmatic-notes-running-dangerous-ai-agents-cloud-vms/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46704650">https://news.ycombinator.com/item?id=46704650</a></p>
<p>Points: 4</p>
<p># Comments: 1</p>
]]></description><pubDate>Wed, 21 Jan 2026 12:21:37 +0000</pubDate><link>https://jakobs.dev/pragmatic-notes-running-dangerous-ai-agents-cloud-vms/</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=46704650</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46704650</guid></item><item><title><![CDATA[New comment by js98 in "Ask HN: Share your personal website"]]></title><description><![CDATA[
<p><a href="https://jakobs.dev" rel="nofollow">https://jakobs.dev</a>
Sharing learnings from engineering work!</p>
]]></description><pubDate>Thu, 15 Jan 2026 07:45:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=46629370</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=46629370</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46629370</guid></item><item><title><![CDATA[Where is Point Nemo? (the point furthest away from land)]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.lukatela.com/pointNemoRevisited/index.html">https://www.lukatela.com/pointNemoRevisited/index.html</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46552266">https://news.ycombinator.com/item?id=46552266</a></p>
<p>Points: 3</p>
<p># Comments: 1</p>
]]></description><pubDate>Fri, 09 Jan 2026 10:28:51 +0000</pubDate><link>https://www.lukatela.com/pointNemoRevisited/index.html</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=46552266</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46552266</guid></item><item><title><![CDATA[New comment by js98 in "ChatGPT Atlas breaks down on simple, scoped repetitive browsing tasks"]]></title><description><![CDATA[
<p>I tested ChatGPT Atlas as an alternative to writing a small scraper for collecting price distribution counts from Dubizzle (UAE Marketplace). It could extract correct numbers for a single filter, but consistently failed to repeat the same simple browsing steps across categories, hallucinating completion or stopping with vague time-constraint explanations. Curious if others have seen similar behavior with LLM-based browsing tools (or have had an alternative work for them)</p>
]]></description><pubDate>Mon, 05 Jan 2026 12:47:24 +0000</pubDate><link>https://news.ycombinator.com/item?id=46498163</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=46498163</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46498163</guid></item><item><title><![CDATA[ChatGPT Atlas breaks down on simple, scoped repetitive browsing tasks]]></title><description><![CDATA[
<p>Article URL: <a href="https://jakobs.dev/chatgpt-atlas-doesnt-have-time-for-me/">https://jakobs.dev/chatgpt-atlas-doesnt-have-time-for-me/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46498158">https://news.ycombinator.com/item?id=46498158</a></p>
<p>Points: 3</p>
<p># Comments: 1</p>
]]></description><pubDate>Mon, 05 Jan 2026 12:46:26 +0000</pubDate><link>https://jakobs.dev/chatgpt-atlas-doesnt-have-time-for-me/</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=46498158</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46498158</guid></item><item><title><![CDATA[New comment by js98 in "Production RAG: what I learned from processing 5M+ documents"]]></title><description><![CDATA[
<p>Similar writeup I did about 1.5 years ago for processing millions of (technical) pages for RAG. Lots has stayed the same it seems<p><a href="https://jakobs.dev/learnings-ingesting-millions-pages-rag-azure/" rel="nofollow">https://jakobs.dev/learnings-ingesting-millions-pages-rag-az...</a></p>
]]></description><pubDate>Mon, 20 Oct 2025 17:01:45 +0000</pubDate><link>https://news.ycombinator.com/item?id=45646254</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=45646254</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45646254</guid></item><item><title><![CDATA[New comment by js98 in "Urban legend: I think there is a world market for maybe five computers"]]></title><description><![CDATA[
<p>This website is unreadable on mobile.</p>
]]></description><pubDate>Fri, 24 Jan 2025 19:12:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=42816116</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=42816116</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42816116</guid></item><item><title><![CDATA[New comment by js98 in "Show HN: I made a puzzle game that gently introduces my favorite math mysteries"]]></title><description><![CDATA[
<p>Nice work! I spent a bit too much time creating a map with 5 colors… ;)</p>
]]></description><pubDate>Thu, 20 Jun 2024 16:36:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=40740587</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=40740587</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40740587</guid></item><item><title><![CDATA[New comment by js98 in "Today's AI Isn't Sentient"]]></title><description><![CDATA[
<p>This seems just like a single argument and counter argument against the sentience (or lack thereof) of AI today. I feel like the article lacks some broader views on the nature of sentience, and is quite narrow in its approach<p>Although, in fairness, I probably wouldn’t make a much better case for either of the sides</p>
]]></description><pubDate>Wed, 29 May 2024 15:37:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=40513099</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=40513099</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40513099</guid></item><item><title><![CDATA[Learnings from ingesting 5M+ technical pages for RAG on Azure]]></title><description><![CDATA[
<p>Article URL: <a href="https://jakobs.dev/learnings-ingesting-millions-pages-rag-azure/">https://jakobs.dev/learnings-ingesting-millions-pages-rag-azure/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=40343522">https://news.ycombinator.com/item?id=40343522</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Mon, 13 May 2024 14:03:57 +0000</pubDate><link>https://jakobs.dev/learnings-ingesting-millions-pages-rag-azure/</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=40343522</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40343522</guid></item><item><title><![CDATA[New comment by js98 in "Silk Helped the Armies of Genghis Khan Conquer Asia"]]></title><description><![CDATA[
<p>Good story, but the writing is annoyingly tough to follow.</p>
]]></description><pubDate>Fri, 03 May 2024 05:24:45 +0000</pubDate><link>https://news.ycombinator.com/item?id=40244349</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=40244349</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40244349</guid></item><item><title><![CDATA[Azure AI 'Hybrid' search gotcha: returning unexpected/unmatching results]]></title><description><![CDATA[
<p>Article URL: <a href="https://jakobs.dev/showcase-azure-ai-hybrid-search-unexpected-results-gotcha/">https://jakobs.dev/showcase-azure-ai-hybrid-search-unexpected-results-gotcha/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=40221797">https://news.ycombinator.com/item?id=40221797</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Wed, 01 May 2024 11:18:22 +0000</pubDate><link>https://jakobs.dev/showcase-azure-ai-hybrid-search-unexpected-results-gotcha/</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=40221797</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40221797</guid></item><item><title><![CDATA[Ask HN: What is the current (Apr. 2024) gold standard of running an LLM locally?]]></title><description><![CDATA[
<p>There are many options and opinions about, what is currently the recommended approach for running an LLM locally (e.g., on my 3090 24Gb)? Are options ‘idiot proof’ yet?</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39893142">https://news.ycombinator.com/item?id=39893142</a></p>
<p>Points: 195</p>
<p># Comments: 88</p>
]]></description><pubDate>Mon, 01 Apr 2024 11:52:41 +0000</pubDate><link>https://news.ycombinator.com/item?id=39893142</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=39893142</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39893142</guid></item><item><title><![CDATA[Show HN: Inject hidden prompt in LLMs using Base64 encoding]]></title><description><![CDATA[
<p>Article URL: <a href="https://jakobs.dev/gpt-hidden-prompt-base64-attack-vector/">https://jakobs.dev/gpt-hidden-prompt-base64-attack-vector/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=38998880">https://news.ycombinator.com/item?id=38998880</a></p>
<p>Points: 5</p>
<p># Comments: 0</p>
]]></description><pubDate>Mon, 15 Jan 2024 09:29:32 +0000</pubDate><link>https://jakobs.dev/gpt-hidden-prompt-base64-attack-vector/</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=38998880</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=38998880</guid></item><item><title><![CDATA[New comment by js98 in "Show HN: Interactive AI Resume/LinkedIn for better networking/job hunting"]]></title><description><![CDATA[
<p>Ha this is great! I built something similar to this for myself a few months back at my personal site at <a href="https://jakobs.dev/chat" rel="nofollow noreferrer">https://jakobs.dev/chat</a></p>
]]></description><pubDate>Mon, 13 Nov 2023 07:16:26 +0000</pubDate><link>https://news.ycombinator.com/item?id=38247795</link><dc:creator>js98</dc:creator><comments>https://news.ycombinator.com/item?id=38247795</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=38247795</guid></item></channel></rss>