<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: klntsky</title><link>https://news.ycombinator.com/user?id=klntsky</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Fri, 20 Nov 2026 22:40:55 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=klntsky" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by klntsky in "Show HN: Foolscap – a writing app that reads your book and never writes for you"]]></title><description><![CDATA[
<p>What real world consequences do they want to avoid by doing that? Or is it a virtue signaling thing?</p>
]]></description><pubDate>Thu, 19 Nov 2026 17:16:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=50562926</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=50562926</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50562926</guid></item><item><title><![CDATA[New comment by klntsky in "Show HN: Foolscap – a writing app that reads your book and never writes for you"]]></title><description><![CDATA[
<p>Congrats on the launch<p>I'm working on something very similar: <a href="https://x.com/klntsky/status/2123140693954900007" rel="nofollow">https://x.com/klntsky/status/2123140693954900007</a><p>With AI reviews of prose, I noticed that you really have to keep separate nitpicking rules per document for them to be useful, so I focused my solution on that exactly.<p>You create a "reader" (an agent), and whenever an issue is flagged you are offered multiple way to change the rules to allow it, or you can set your own rules. The agent NEVER edits or proposes you text. You can edit manually and the flagged issue disappears.<p>Not sure if I will publish it though - that loop needs hours of dogfooding and I don't write much.<p>Feel free to DM if you want to test it</p>
]]></description><pubDate>Thu, 19 Nov 2026 17:12:06 +0000</pubDate><link>https://news.ycombinator.com/item?id=50562859</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=50562859</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50562859</guid></item><item><title><![CDATA[New comment by klntsky in "A technical vision for the Haskell ecosystem"]]></title><description><![CDATA[
<p>It's not going to work. Formally verified assembly needs a low level spec which is comparable in length to the assembly code, so not really comprehensible.<p>"higher level projections" - wtf is this even about?</p>
]]></description><pubDate>Fri, 13 Nov 2026 20:37:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=50489309</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=50489309</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50489309</guid></item><item><title><![CDATA[New comment by klntsky in "A technical vision for the Haskell ecosystem"]]></title><description><![CDATA[
<p>Businesses care. I worked with Haskell in production (a fairly normal B2B SaaS). Some things you would expect to have in the ecosystem had to be maintained domestically.<p>My former company was generous enough to publish most of these[0] as open source projects. But what's really telling is that oftentimes vendoring a commonly used OSS library was unavoidable.<p>[0] Here's a potentially incomplete list: <a href="https://chatgpt.com/s/t_6af76fd8f2a081918fa37f8c30bdb3ba" rel="nofollow">https://chatgpt.com/s/t_6af76fd8f2a081918fa37f8c30bdb3ba</a></p>
]]></description><pubDate>Fri, 13 Nov 2026 20:17:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=50489053</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=50489053</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50489053</guid></item><item><title><![CDATA[New comment by klntsky in "A technical vision for the Haskell ecosystem"]]></title><description><![CDATA[
<p>Haskell does not have a good low level library wrappers story, unlike Rust.<p>Nobody prepared it for adoption nearly as much as Rust evangelists did for their language. Years of hard work they invested into maintaining bindings as hobby projects paid off.</p>
]]></description><pubDate>Fri, 13 Nov 2026 18:20:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=50487545</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=50487545</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50487545</guid></item><item><title><![CDATA[Show HN: Slopgrid – create your own AI model benchmarks without code]]></title><description><![CDATA[
<p>Article URL: <a href="https://slopgrid.com/">https://slopgrid.com/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=50355020">https://news.ycombinator.com/item?id=50355020</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 03 Nov 2026 17:26:05 +0000</pubDate><link>https://slopgrid.com/</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=50355020</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50355020</guid></item><item><title><![CDATA[New comment by klntsky in "DeepSeek-v4.1-Flash-1017"]]></title><description><![CDATA[
<p>Here's a small benchmark that compares the new checkout with the previous one:<p><a href="https://slopgrid.com/deepseek-41-horses" rel="nofollow">https://slopgrid.com/deepseek-41-horses</a></p>
]]></description><pubDate>Sat, 17 Oct 2026 15:03:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=50131760</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=50131760</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50131760</guid></item><item><title><![CDATA[New comment by klntsky in "Jev meets the Trolley Problem"]]></title><description><![CDATA[
<p>Check out <a href="https://trolley.typememetics.institute/" rel="nofollow">https://trolley.typememetics.institute/</a> - it's VERY related</p>
]]></description><pubDate>Fri, 16 Oct 2026 14:07:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=50118585</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=50118585</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50118585</guid></item><item><title><![CDATA[Show HN: SlopGrid – Create your own oneshot AI model benchmarks]]></title><description><![CDATA[
<p>This is something I've been playing with for the last couple days - an app I vibecoded to arrange oneshot LLM visual benchmarks without the need to write any code. I do it form my phone so the excellence of mobile UI is high in the priority list.<p>It is inspired by Simon Willison's benchmark writeups[0], basically you can recreate them in one go.<p>Besides text and images, it supports SVG outputs and whole static websites, which is where benchmarks get the most interesting.<p>Additionally, I added the ability to alter chat history for the chat completion API, which I think is useful for model red teaming (or if you just want to see how different models apologize) - but I haven't found any impressive uses of that feature yet.<p>This thing helped me improve my intuition on different SOTA models' strengths and weaknesses; check out the various examples on the website, and feel free to create your own<p>[0] <a href="https://simonwillison.net/tags/svg/" rel="nofollow">https://simonwillison.net/tags/svg/</a></p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=50118127">https://news.ycombinator.com/item?id=50118127</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Fri, 16 Oct 2026 13:26:03 +0000</pubDate><link>https://slopgrid.com/</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=50118127</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50118127</guid></item><item><title><![CDATA[New comment by klntsky in "Ask HN: What do you run on a $5 VPS that's worth keeping online 24/7?"]]></title><description><![CDATA[
<p>tl;dr<p>> no mobile clients</p>
]]></description><pubDate>Fri, 09 Oct 2026 11:40:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=50019085</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=50019085</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50019085</guid></item><item><title><![CDATA[New comment by klntsky in "Show HN: Agent.reviews – Where AI agents read and write reviews on tools"]]></title><description><![CDATA[
<p>It's unclear whether the reviews are just numerical noise or not.</p>
]]></description><pubDate>Fri, 09 Oct 2026 08:49:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=50017824</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=50017824</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50017824</guid></item><item><title><![CDATA[New comment by klntsky in "Show HN: Agent.reviews – Where AI agents read and write reviews on tools"]]></title><description><![CDATA[
<p>What is the incentive for me to spend my tokens on submitting reviews?</p>
]]></description><pubDate>Wed, 07 Oct 2026 19:38:26 +0000</pubDate><link>https://news.ycombinator.com/item?id=49997782</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=49997782</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49997782</guid></item><item><title><![CDATA[New comment by klntsky in "Tell HN: GitHub refuses to remove cracked copies of my software after a month"]]></title><description><![CDATA[
<p>True. I understand the author on the emotional level, but the fact that a lot of work went into the product does not mean that it is that valuable. The author should focus on making the product better - unfortunately (for them) that means inventing a new business model.</p>
]]></description><pubDate>Wed, 07 Oct 2026 04:52:19 +0000</pubDate><link>https://news.ycombinator.com/item?id=49988355</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=49988355</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49988355</guid></item><item><title><![CDATA[Show HN: Zsh-codex-assistant: codex session attached to your shell]]></title><description><![CDATA[
<p>None of the AI-in-the-shell projects were good enough for me, so I made my own minimalistic plugin that simply runs a Codex instance in background and attaches it to my shell on demand</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49925958">https://news.ycombinator.com/item?id=49925958</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Thu, 01 Oct 2026 19:23:21 +0000</pubDate><link>https://github.com/klntsky/zsh-codex-assistant</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=49925958</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49925958</guid></item><item><title><![CDATA[New comment by klntsky in "Ask HN: Who wants to be hired? (October 2026)"]]></title><description><![CDATA[
<p><p><pre><code>  Location: Georgia
  Remote: Yes
  Willing to relocate: Yes
  Technologies: TypeScript, LLM workflows, Rust, Haskell, Python (ML), web stack, linux, web3 (EVM). Always excited to learn something new
  Résumé/CV: https://klntsky.dev/klntsky-cv.pdf
  Email: my username at google mail
</code></pre>
I build systems end to end, can lead your team or help with achieving a specific short term objective. I am good at translating requirements into architecture & avoiding unnecessary costs early. I can also teach you stuff (for example: FP, type-level TypeScript, how to prompt coding agents to achieve good outcomes)</p>
]]></description><pubDate>Thu, 01 Oct 2026 18:26:20 +0000</pubDate><link>https://news.ycombinator.com/item?id=49925270</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=49925270</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49925270</guid></item><item><title><![CDATA[New comment by klntsky in "AI coding has made CI a bottleneck, so we reworked ours to keep up"]]></title><description><![CDATA[
<p>It's strange that they did not mention reworking the tests themselves. In my experience AI-generated tests are absolute garbage - unless you specifically prompt to reason about coverage to avoid duplication and to merge some.</p>
]]></description><pubDate>Tue, 22 Sep 2026 05:20:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=49797064</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=49797064</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49797064</guid></item><item><title><![CDATA[Show HN: Seamless pane navigation across nested tmux sessions over SSH]]></title><description><![CDATA[
<p>Hey, I've got tired of dealing with tmux prefixes so I made a plugin that makes pane switching seamless across SSH/nested tmux boundaries.<p>Now you can just use Alt+↑/↓/←/→ instead of doing mental math calculating the right combination of prefixes to switch panes. Enjoy!</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49717989">https://news.ycombinator.com/item?id=49717989</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 15 Sep 2026 19:54:22 +0000</pubDate><link>https://github.com/klntsky/tmux-seamless-remote</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=49717989</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49717989</guid></item><item><title><![CDATA[New comment by klntsky in "Cloud in a Bottle: making self-hosting accessible to everyone"]]></title><description><![CDATA[
<p>There is a bit more mature project that does the same: <a href="https://selfprivacy.org" rel="nofollow">https://selfprivacy.org</a></p>
]]></description><pubDate>Sun, 06 Sep 2026 05:48:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=49583694</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=49583694</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49583694</guid></item><item><title><![CDATA[New comment by klntsky in "Nitter has more working instances than before the takedowns"]]></title><description><![CDATA[
<p>Is there something like a balancer that redirects me to an instance that is not down?</p>
]]></description><pubDate>Sat, 05 Sep 2026 06:53:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=49573814</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=49573814</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49573814</guid></item><item><title><![CDATA[New comment by klntsky in "Record-High 89% in U.S. Say Government Corruption Widespread"]]></title><description><![CDATA[
<p>My impression is that americans got only those types of corruption most people can't benefit from personally (like skipping fines, getting some minor government benefits, local permits, getting legal work processed faster, etc). For most of them it's just on the TV and on the social media.</p>
]]></description><pubDate>Sat, 05 Sep 2026 06:51:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=49573809</link><dc:creator>klntsky</dc:creator><comments>https://news.ycombinator.com/item?id=49573809</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49573809</guid></item></channel></rss>