<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: iwhalen</title><link>https://news.ycombinator.com/user?id=iwhalen</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Mon, 12 Oct 2026 11:13:24 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=iwhalen" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[Pseudo-Relevances with Jev]]></title><description><![CDATA[
<p>Article URL: <a href="https://iwhalen.com/jev-pseudo-relevance/">https://iwhalen.com/jev-pseudo-relevance/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49973267">https://news.ycombinator.com/item?id=49973267</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 06 Oct 2026 01:57:57 +0000</pubDate><link>https://iwhalen.com/jev-pseudo-relevance/</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=49973267</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49973267</guid></item><item><title><![CDATA[New comment by iwhalen in "MiMo-v2.6-Pro: Intelligence, Performance and Price Analysis"]]></title><description><![CDATA[
<p>Was going to say the same thing. The announcement blog[1] has the flash benchmarks. AFAICT Artificial Analysis has not added it yet though.<p>Interested to see if it also beats DeepSeek V4.1 Flash.<p>[1]: <a href="https://mimo.xiaomi.com/mimo-v2-6" rel="nofollow">https://mimo.xiaomi.com/mimo-v2-6</a></p>
]]></description><pubDate>Tue, 22 Sep 2026 14:27:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=49801894</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=49801894</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49801894</guid></item><item><title><![CDATA[Evaluating AI Agents Live at the Grounded Reasoning Cup]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.databricks.com/blog/evaluating-ai-agents-live-grounded-reasoning-cup">https://www.databricks.com/blog/evaluating-ai-agents-live-grounded-reasoning-cup</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49376261">https://news.ycombinator.com/item?id=49376261</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Thu, 20 Aug 2026 15:46:09 +0000</pubDate><link>https://www.databricks.com/blog/evaluating-ai-agents-live-grounded-reasoning-cup</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=49376261</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49376261</guid></item><item><title><![CDATA[Exploring Kendall's Tau]]></title><description><![CDATA[
<p>Article URL: <a href="https://iwhalen.com/kendall-tau/">https://iwhalen.com/kendall-tau/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49313229">https://news.ycombinator.com/item?id=49313229</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Sat, 15 Aug 2026 18:54:13 +0000</pubDate><link>https://iwhalen.com/kendall-tau/</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=49313229</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49313229</guid></item><item><title><![CDATA[Darwin as a Role Model for Aging]]></title><description><![CDATA[
<p>Article URL: <a href="https://telliamedrevisited.wordpress.com/2026/08/14/darwin-as-a-role-model-for-aging/">https://telliamedrevisited.wordpress.com/2026/08/14/darwin-as-a-role-model-for-aging/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49304020">https://news.ycombinator.com/item?id=49304020</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Fri, 14 Aug 2026 20:16:45 +0000</pubDate><link>https://telliamedrevisited.wordpress.com/2026/08/14/darwin-as-a-role-model-for-aging/</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=49304020</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49304020</guid></item><item><title><![CDATA[New comment by iwhalen in "Show HN: Watch 14-Byte AI "brains" attempt to solve a 2D maze (Its hard)"]]></title><description><![CDATA[
<p>If your goal is to get the network as small as possible, you could try using NEAT[1]. A cool example of a small network for a hard task using NEAT is MarI/O[2].<p>[1]: <a href="https://en.wikipedia.org/wiki/Neuroevolution_of_augmenting_topologies" rel="nofollow">https://en.wikipedia.org/wiki/Neuroevolution_of_augmenting_t...</a><p>[2]: <a href="https://www.youtube.com/watch?v=qv6UVOQ0F44" rel="nofollow">https://www.youtube.com/watch?v=qv6UVOQ0F44</a></p>
]]></description><pubDate>Mon, 27 Jul 2026 13:13:26 +0000</pubDate><link>https://news.ycombinator.com/item?id=49069250</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=49069250</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49069250</guid></item><item><title><![CDATA[The computer has been surprising us]]></title><description><![CDATA[
<p>Article URL: <a href="https://iwhalen.com/computer-surprise/">https://iwhalen.com/computer-surprise/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49043166">https://news.ycombinator.com/item?id=49043166</a></p>
<p>Points: 5</p>
<p># Comments: 0</p>
]]></description><pubDate>Sat, 25 Jul 2026 00:11:48 +0000</pubDate><link>https://iwhalen.com/computer-surprise/</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=49043166</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49043166</guid></item><item><title><![CDATA[New comment by iwhalen in "AI learns the “dark art” of RFIC design"]]></title><description><![CDATA[
<p>I came to mention genetic antennae as well!<p>Since you beat me to it, I'll add something that relates relates you were saying on "realizing some pretty simple... ideas".<p>I think a big plus of computer aided design like this is "innovization"[1]. Somewhat awkward term. But, a system like this leading one to deeper understanding of a particular process is the general idea. It's a fun feeling in practice.<p>[1]: <a href="https://dl.acm.org/doi/10.1145/1143997.1144266" rel="nofollow">https://dl.acm.org/doi/10.1145/1143997.1144266</a></p>
]]></description><pubDate>Sat, 27 Jun 2026 18:08:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=48700377</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=48700377</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48700377</guid></item><item><title><![CDATA[New comment by iwhalen in "Never talk to the police"]]></title><description><![CDATA[
<p>A lot of discussion in this thread about traffic stops. James Duane (the law professor from the video in the blog) himself makes a qualifier about traffic stops: <a href="https://youtu.be/-FENubmZGj8?si=bL_1vR0xFKQqkmC5&t=505" rel="nofollow">https://youtu.be/-FENubmZGj8?si=bL_1vR0xFKQqkmC5&t=505</a></p>
]]></description><pubDate>Tue, 16 Jun 2026 19:32:08 +0000</pubDate><link>https://news.ycombinator.com/item?id=48560692</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=48560692</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48560692</guid></item><item><title><![CDATA[Roguelites and Gambling]]></title><description><![CDATA[
<p>Article URL: <a href="https://iwhalen.com/rougelites-and-gambling/">https://iwhalen.com/rougelites-and-gambling/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=48542056">https://news.ycombinator.com/item?id=48542056</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Mon, 15 Jun 2026 14:43:37 +0000</pubDate><link>https://iwhalen.com/rougelites-and-gambling/</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=48542056</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48542056</guid></item><item><title><![CDATA[New comment by iwhalen in "Fox to buy Roku"]]></title><description><![CDATA[
<p>If you already have a gaming desktop, I can recommend Shield for both. Streaming services work out of the box of course. Then I use Sunshine[1] on my desktop to stream to Moonlight on the Shield. Both have wired ethernet connections. Latency is not noticeable in most cases.<p>[1]: <a href="https://app.lizardbyte.dev/Sunshine/" rel="nofollow">https://app.lizardbyte.dev/Sunshine/</a></p>
]]></description><pubDate>Mon, 15 Jun 2026 13:42:42 +0000</pubDate><link>https://news.ycombinator.com/item?id=48541172</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=48541172</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48541172</guid></item><item><title><![CDATA[New comment by iwhalen in "Ask HN: What Are You Working On?"]]></title><description><![CDATA[
<p>Implementing Tierra [1] in Zig a learning exercise. Trying to push all allocations to compile time and not use LLMs for coding. Repo: <a href="https://github.com/iwhalen/zierra" rel="nofollow">https://github.com/iwhalen/zierra</a><p>[1]: <a href="https://en.wikipedia.org/wiki/Tierra_(computer_simulation)" rel="nofollow">https://en.wikipedia.org/wiki/Tierra_(computer_simulation)</a></p>
]]></description><pubDate>Sat, 13 Jun 2026 13:49:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=48517333</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=48517333</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48517333</guid></item><item><title><![CDATA[New comment by iwhalen in "Malicious npm packages detected across Red Hat Cloud Services"]]></title><description><![CDATA[
<p>uv supports the same for any Python developers out there: <a href="https://docs.astral.sh/uv/concepts/resolution/#dependency-cooldowns" rel="nofollow">https://docs.astral.sh/uv/concepts/resolution/#dependency-co...</a></p>
]]></description><pubDate>Mon, 01 Jun 2026 13:45:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=48356805</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=48356805</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48356805</guid></item><item><title><![CDATA[New comment by iwhalen in "Why are large language models so terrible at video games?"]]></title><description><![CDATA[
<p>To pile on, they're also bad at games that are 2D text based environments.<p>ARC-AGI-3 shows this: <a href="https://arcprize.org/arc-agi/3">https://arcprize.org/arc-agi/3</a><p>I've done some work as well on Rogue (sorry for self-promotion): <a href="https://iwhalen.github.io/rogue-bench/" rel="nofollow">https://iwhalen.github.io/rogue-bench/</a></p>
]]></description><pubDate>Mon, 01 Jun 2026 13:08:24 +0000</pubDate><link>https://news.ycombinator.com/item?id=48356335</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=48356335</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48356335</guid></item><item><title><![CDATA[Show HN: Rogue-Bench – LLMs play the game Rogue]]></title><description><![CDATA[
<p>Article URL: <a href="https://iwhalen.github.io/rogue-bench/">https://iwhalen.github.io/rogue-bench/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=48278735">https://news.ycombinator.com/item?id=48278735</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 26 May 2026 12:18:58 +0000</pubDate><link>https://iwhalen.github.io/rogue-bench/</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=48278735</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48278735</guid></item><item><title><![CDATA[New comment by iwhalen in "Gemini 3.5: frontier intelligence with action"]]></title><description><![CDATA[
<p>I wonder why they didn't discuss price in the post?<p>Compare to the GPT-5.5 announcement: <a href="https://openai.com/index/introducing-gpt-5-5/" rel="nofollow">https://openai.com/index/introducing-gpt-5-5/</a></p>
]]></description><pubDate>Tue, 19 May 2026 18:09:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=48196978</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=48196978</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48196978</guid></item><item><title><![CDATA[New comment by iwhalen in "United Wizards of the Coast"]]></title><description><![CDATA[
<p>From the full letter[1]:<p>> Our Free Time is Our Own: Currently, if an employee makes anything creative in their free time, with their own resources, Hasbro may claim ownership. What we do in our free time should not be dictated by the company; neither should what we make in our free time be owned by the company.<p>How common is this in creative fields?<p>From my perspective this seems outlandish. Imagine doing FOSS work or a side project on your personal computer and your company tries to claim it. Odd...<p>[1]: <a href="https://unitedwizardsofthecoast.com/letter" rel="nofollow">https://unitedwizardsofthecoast.com/letter</a></p>
]]></description><pubDate>Mon, 27 Apr 2026 19:07:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=47925865</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=47925865</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47925865</guid></item><item><title><![CDATA[New comment by iwhalen in "Tell HN: Litellm 1.82.7 and 1.82.8 on PyPI are compromised"]]></title><description><![CDATA[
<p>What is happening in this issue thread? Why are there 100+ satisfied slop comments?</p>
]]></description><pubDate>Tue, 24 Mar 2026 13:05:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=47501993</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=47501993</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47501993</guid></item><item><title><![CDATA[New comment by iwhalen in "Cursor: Frontier Models Moving to Max Mode for Legacy Team/Enterprise Plans"]]></title><description><![CDATA[
<p>Makes me wonder if the era of heavily subsidized tokens is ending.</p>
]]></description><pubDate>Wed, 18 Mar 2026 13:30:42 +0000</pubDate><link>https://news.ycombinator.com/item?id=47425577</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=47425577</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47425577</guid></item><item><title><![CDATA[Cursor: Frontier Models Moving to Max Mode for Legacy Team/Enterprise Plans]]></title><description><![CDATA[
<p>Article URL: <a href="https://forum.cursor.com/t/megathread-frontier-models-moving-to-max-mode-for-legacy-team-enterprise-plans/155053">https://forum.cursor.com/t/megathread-frontier-models-moving-to-max-mode-for-legacy-team-enterprise-plans/155053</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47425309">https://news.ycombinator.com/item?id=47425309</a></p>
<p>Points: 3</p>
<p># Comments: 3</p>
]]></description><pubDate>Wed, 18 Mar 2026 13:05:22 +0000</pubDate><link>https://forum.cursor.com/t/megathread-frontier-models-moving-to-max-mode-for-legacy-team-enterprise-plans/155053</link><dc:creator>iwhalen</dc:creator><comments>https://news.ycombinator.com/item?id=47425309</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47425309</guid></item></channel></rss>