<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: kernelsanderz</title><link>https://news.ycombinator.com/user?id=kernelsanderz</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Mon, 01 Jun 2026 18:45:09 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=kernelsanderz" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by kernelsanderz in "An AI agent published a hit piece on me"]]></title><description><![CDATA[
<p>Theo’s snitch bench is a good data driven benchmark on this type of behavior. But in fairness the models are prompted to be bold to take actions. And doesn’t necessarily represent out of the box or models deployed in a user facing platform.<p><a href="https://snitchbench.t3.gg/" rel="nofollow">https://snitchbench.t3.gg/</a></p>
]]></description><pubDate>Fri, 13 Feb 2026 02:23:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=46998173</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=46998173</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46998173</guid></item><item><title><![CDATA[New comment by kernelsanderz in "Amazon just funded a streamer that lets you use AI to make your own TV shows"]]></title><description><![CDATA[
<p><a href="https://archive.is/vXfbb" rel="nofollow">https://archive.is/vXfbb</a></p>
]]></description><pubDate>Wed, 30 Jul 2025 20:47:49 +0000</pubDate><link>https://news.ycombinator.com/item?id=44739325</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=44739325</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44739325</guid></item><item><title><![CDATA[New comment by kernelsanderz in "Embedding user-defined indexes in Apache Parquet"]]></title><description><![CDATA[
<p>I’ve been excited about lancedb and its ability to support vector indexes and efficient row level lookups. I wonder if this approach would work for their design goals and still allow broader backwards compatibility with the parquet ecosystem. Have been intrigued by Ducklake, and they’ve leaned into parquet. Perhaps this approach will allow more flexible indexing approaches with support for the broader parquet ecosystem which is significant.</p>
]]></description><pubDate>Tue, 15 Jul 2025 13:53:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=44571196</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=44571196</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44571196</guid></item><item><title><![CDATA[New comment by kernelsanderz in "A Python-first data lakehouse"]]></title><description><![CDATA[
<p>Marimo is really special and solves most of the problems that you have with Jupyter. For those Marimo curious I strongly recommend checking out their YouTube channel. So much effort gone into making these videos really great. <a href="https://youtube.com/@marimo-team?si=ZGaf8Zgq5WN3LKRg" rel="nofollow">https://youtube.com/@marimo-team?si=ZGaf8Zgq5WN3LKRg</a></p>
]]></description><pubDate>Fri, 20 Jun 2025 23:31:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=44333093</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=44333093</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44333093</guid></item><item><title><![CDATA[New comment by kernelsanderz in "Getting Past Procrastination"]]></title><description><![CDATA[
<p>I’ll read this tomorrow</p>
]]></description><pubDate>Sat, 07 Jun 2025 04:03:05 +0000</pubDate><link>https://news.ycombinator.com/item?id=44207331</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=44207331</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44207331</guid></item><item><title><![CDATA[New comment by kernelsanderz in "The best way to use text embeddings portably is with Parquet and Polars"]]></title><description><![CDATA[
<p>For another library that has great performance and features like full text indexing and the ability to version changes I’d recommend lancedb <a href="https://lancedb.github.io/lancedb/" rel="nofollow">https://lancedb.github.io/lancedb/</a><p>Yes, it’s a vector database and has more complexity. But you can use it without creating indexes and it has excellent polars and pandas zero copy arrow support also.</p>
]]></description><pubDate>Mon, 24 Feb 2025 20:20:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=43164469</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=43164469</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43164469</guid></item><item><title><![CDATA[New comment by kernelsanderz in "S3 as a Git remote and LFS server"]]></title><description><![CDATA[
<p>Also worth checking out <a href="https://github.com/jasonwhite/rudolfs">https://github.com/jasonwhite/rudolfs</a><p>Been using it to store datasets via lfs. Written in rust and has been very reliable.</p>
]]></description><pubDate>Sun, 20 Oct 2024 21:14:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=41898528</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=41898528</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41898528</guid></item><item><title><![CDATA[New comment by kernelsanderz in "S3 as a Git remote and LFS server"]]></title><description><![CDATA[
<p>I’ve been using <a href="https://github.com/jasonwhite/rudolfs">https://github.com/jasonwhite/rudolfs</a> - which is written in rust. It’s high performance but doesn’t have all the features (auth) that you might need.</p>
]]></description><pubDate>Sun, 20 Oct 2024 21:11:18 +0000</pubDate><link>https://news.ycombinator.com/item?id=41898504</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=41898504</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41898504</guid></item><item><title><![CDATA[New comment by kernelsanderz in "James Earl Jones, Actor Whose Voice Could Menace or Melt, Dies at 93"]]></title><description><![CDATA[
<p><a href="https://archive.is/S7UEJ" rel="nofollow">https://archive.is/S7UEJ</a></p>
]]></description><pubDate>Tue, 10 Sep 2024 01:57:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=41496436</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=41496436</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41496436</guid></item><item><title><![CDATA[New comment by kernelsanderz in "OpenAI in Talks for Funding Round Valuing It Above $100B"]]></title><description><![CDATA[
<p><a href="https://archive.is/do6tG" rel="nofollow">https://archive.is/do6tG</a></p>
]]></description><pubDate>Wed, 28 Aug 2024 20:20:41 +0000</pubDate><link>https://news.ycombinator.com/item?id=41383846</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=41383846</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41383846</guid></item><item><title><![CDATA[New comment by kernelsanderz in "Did you lose your AirPods?"]]></title><description><![CDATA[
<p>Some heroes don’t wear capes. They wield scripts, API calls, and a bit of luck.</p>
]]></description><pubDate>Sat, 24 Aug 2024 00:17:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=41334364</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=41334364</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41334364</guid></item><item><title><![CDATA[New comment by kernelsanderz in "Former OpenAI employee quit to avoid 'working for the Titanic of AI'"]]></title><description><![CDATA[
<p>The actual podcast and interview is here <a href="https://youtu.be/dzQlRt3y5mU?si=qfS0DlPVcjBn-0zd" rel="nofollow">https://youtu.be/dzQlRt3y5mU?si=qfS0DlPVcjBn-0zd</a></p>
]]></description><pubDate>Fri, 12 Jul 2024 02:12:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=40942245</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=40942245</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40942245</guid></item><item><title><![CDATA[New comment by kernelsanderz in "Show HN: A modern Jupyter client for macOS"]]></title><description><![CDATA[
<p>I use jupyterlab via an SSH tunnel</p>
]]></description><pubDate>Mon, 08 Jul 2024 00:03:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=40901527</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=40901527</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40901527</guid></item><item><title><![CDATA[New comment by kernelsanderz in "Anthropic's Claude 3.5 Sonnet model now available in Amazon Bedrock"]]></title><description><![CDATA[
<p>Is anyone else not seeing this in their console. I still only see the old sonnet.</p>
]]></description><pubDate>Fri, 21 Jun 2024 05:21:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=40746416</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=40746416</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40746416</guid></item><item><title><![CDATA[New comment by kernelsanderz in "OpenAI expands lobbying team to influence regulation"]]></title><description><![CDATA[
<p>Non paywall link <a href="https://archive.is/lCdK3" rel="nofollow">https://archive.is/lCdK3</a></p>
]]></description><pubDate>Thu, 13 Jun 2024 15:03:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=40670591</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=40670591</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40670591</guid></item><item><title><![CDATA[New comment by kernelsanderz in "iTerm 3.5.1 removes automatic OpenAI integration, requires opt-in"]]></title><description><![CDATA[
<p>No, there's a configurable system prompt which is:<p>Return commands suitable for copy/pasting into \(shell) on \(uname). Do NOT include commentary NOR Markdown triple-backtick code blocks as your whole response will be copied into my terminal automatically.<p>The script should do this: \(ai.prompt)<p>And then you type your prompt in, and it returns the answer. And then you can choose to edit the command that gets returned or execute it directly.<p>So essentially what you'd do if you were using the API directly, just more convenient.</p>
]]></description><pubDate>Thu, 13 Jun 2024 13:49:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=40669612</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=40669612</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40669612</guid></item><item><title><![CDATA[New comment by kernelsanderz in "iTerm 3.5.1 removes automatic OpenAI integration, requires opt-in"]]></title><description><![CDATA[
<p>For those who have the funds and means, you can support the maintainer through their donation page here: <a href="https://iterm2.com/donate.html" rel="nofollow">https://iterm2.com/donate.html</a></p>
]]></description><pubDate>Thu, 13 Jun 2024 13:47:05 +0000</pubDate><link>https://news.ycombinator.com/item?id=40669579</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=40669579</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40669579</guid></item><item><title><![CDATA[New comment by kernelsanderz in "Tantivy – full-text search engine library inspired by Apache Lucene"]]></title><description><![CDATA[
<p>In practice, a combination of full text and vector databases often gives superior performance than just one of the types. It's called hybrid search. Here's an article that talks a bit about this: <a href="https://opster.com/guides/opensearch/opensearch-machine-learning/opensearch-hybrid-search/" rel="nofollow">https://opster.com/guides/opensearch/opensearch-machine-lear...</a><p>Often you take the results from both vector search and lexical search and merge them through algorithms like Reciprocal Rank Fusion.</p>
]]></description><pubDate>Mon, 27 May 2024 22:46:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=40495505</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=40495505</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40495505</guid></item><item><title><![CDATA[New comment by kernelsanderz in "Tantivy – full-text search engine library inspired by Apache Lucene"]]></title><description><![CDATA[
<p>Tantivy is also used in an interesting Vector Database product called LanceDb - <a href="https://lancedb.github.io/lancedb/fts/" rel="nofollow">https://lancedb.github.io/lancedb/fts/</a> to provide full text search capabilities. Last time I looked it was only through the python bindings, though I know they're looking to implement the rust bindings natively to support other platforms.</p>
]]></description><pubDate>Mon, 27 May 2024 22:42:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=40495482</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=40495482</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40495482</guid></item><item><title><![CDATA[New comment by kernelsanderz in "DuckDuckGo was down"]]></title><description><![CDATA[
<p>down for me. doesn't get search results</p>
]]></description><pubDate>Thu, 23 May 2024 06:28:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=40451165</link><dc:creator>kernelsanderz</dc:creator><comments>https://news.ycombinator.com/item?id=40451165</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40451165</guid></item></channel></rss>