<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: MasterScrat</title><link>https://news.ycombinator.com/user?id=MasterScrat</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Sat, 21 Nov 2026 20:39:28 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=MasterScrat" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by MasterScrat in "Jev-Leftpad"]]></title><description><![CDATA[
<p>I’d be curious to see how reliable it is in practice, it sounds like a cheap and (hopefully) easy benchmark</p>
]]></description><pubDate>Mon, 21 Sep 2026 11:50:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=49786066</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=49786066</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49786066</guid></item><item><title><![CDATA[New comment by MasterScrat in "Bend 2 and the Vibe-Coding Trap"]]></title><description><![CDATA[
<p>I fully agree with this quote, it really crystallized something i felt but couldn’t put into words.</p>
]]></description><pubDate>Fri, 18 Sep 2026 14:36:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=49755036</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=49755036</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49755036</guid></item><item><title><![CDATA[New comment by MasterScrat in "Your intellectual fly is open when you use an LLM to author a post (2025)"]]></title><description><![CDATA[
<p>I don’t think it’s a good analogy.<p>Having your fly open is a harmless mistake that has little impact on your peers. People may or may not mention it to you but it’s not something they’ll hold against you.<p>Posting LLM slop under your name is a deliberate act. You decide to damage your message by taking a shortcut.<p>The obvious analogy was speaking while chewing.</p>
]]></description><pubDate>Sun, 06 Sep 2026 22:18:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=49591499</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=49591499</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49591499</guid></item><item><title><![CDATA[New comment by MasterScrat in "Karpathy’s Pelican"]]></title><description><![CDATA[
<p>I don't think this is bad. I want tools that deliver what I envision. The fact that given the same input, I get similar output is something I would rather consider a positive. The imagination should come from the human at the wheel.</p>
]]></description><pubDate>Sun, 02 Aug 2026 18:43:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=49147145</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=49147145</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49147145</guid></item><item><title><![CDATA[New comment by MasterScrat in "Generative AI floods and dilutes the market for books"]]></title><description><![CDATA[
<p>> he just read out something like "You're right to push back on this. If you'd like, I can reword this..." and didn't even flinch<p>Can you please provide sources for this? My understanding is that he used the expression "I appreciate the pushback" in a context in which it made sense (responding to criticism), and people flagged that it sounded LLM-like.<p>There’s some poetry seeing people complaining about how LLM provide unreliable information in a confident tone, who then go on to share incorrect information in a confident tone.</p>
]]></description><pubDate>Sun, 02 Aug 2026 13:27:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=49144517</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=49144517</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49144517</guid></item><item><title><![CDATA[New comment by MasterScrat in "Generative AI floods and dilutes the market for books"]]></title><description><![CDATA[
<p>I don’t see how this market ("mostly" AI-written books) won’t get arbitraged to death with the cheaper and open-weight models that keep coming out</p>
]]></description><pubDate>Sun, 02 Aug 2026 11:46:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=49143561</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=49143561</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49143561</guid></item><item><title><![CDATA[Damned if you do and damned if you don't: Can NOT using AI amount to negligence?]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.judiciary.uk/speech-by-the-master-of-the-rolls-to-the-professional-negligence-bar-association/">https://www.judiciary.uk/speech-by-the-master-of-the-rolls-to-the-professional-negligence-bar-association/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49143201">https://news.ycombinator.com/item?id=49143201</a></p>
<p>Points: 2</p>
<p># Comments: 1</p>
]]></description><pubDate>Sun, 02 Aug 2026 10:48:46 +0000</pubDate><link>https://www.judiciary.uk/speech-by-the-master-of-the-rolls-to-the-professional-negligence-bar-association/</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=49143201</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49143201</guid></item><item><title><![CDATA[New comment by MasterScrat in "Claude Opus 5"]]></title><description><![CDATA[
<p>Damn the pelican guy can’t get no sleep</p>
]]></description><pubDate>Fri, 24 Jul 2026 21:17:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=49041693</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=49041693</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49041693</guid></item><item><title><![CDATA[New comment by MasterScrat in "Most Americans say "not in my backyard" to AI data centers"]]></title><description><![CDATA[
<p>> xAI is renting out their DC space. I think I read that Google is, too.<p>You’re probably thinking of Meta. Google renting GPUs is a less recent phenomenon, to say the least.</p>
]]></description><pubDate>Wed, 22 Jul 2026 17:26:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=49010271</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=49010271</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49010271</guid></item><item><title><![CDATA[New comment by MasterScrat in "OpenAI and Hugging Face address security incident during model evaluation"]]></title><description><![CDATA[
<p>From HF:<p>> The practical lesson for defenders: have a capable model you can run on your own infrastructure vetted and ready before an incident, both to avoid guardrail lockout and to keep attacker data and credentials from leaving your environment.<p>This sounds like a reasonable measure. Any recommendations regarding the most suitable models for this? GLM? Kimi?</p>
]]></description><pubDate>Wed, 22 Jul 2026 09:49:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=49004178</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=49004178</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49004178</guid></item><item><title><![CDATA[New comment by MasterScrat in "Nvidia, CoreWeave, and Nebius: Inside the Circular Financing of the GPU Boom"]]></title><description><![CDATA[
<p>How is what you describe "YC-ish"?</p>
]]></description><pubDate>Sun, 12 Jul 2026 10:39:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=48880009</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=48880009</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48880009</guid></item><item><title><![CDATA[New comment by MasterScrat in "MIRA: Multiplayer Interactive World Models Trained on Rocket League"]]></title><description><![CDATA[
<p>Playing backward is actually a great idea for testing generalization</p>
]]></description><pubDate>Sat, 11 Jul 2026 08:18:47 +0000</pubDate><link>https://news.ycombinator.com/item?id=48869972</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=48869972</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48869972</guid></item><item><title><![CDATA[New comment by MasterScrat in "MIRA: Multiplayer Interactive World Models Trained on Rocket League"]]></title><description><![CDATA[
<p>Would be a great idea to see how much we could manipulate the latent space and whether it has some internal structure w.r.t the physical state. I guess the only unknown is how the world model would show robustness to latent states that are transformed through this network</p>
]]></description><pubDate>Thu, 09 Jul 2026 05:34:27 +0000</pubDate><link>https://news.ycombinator.com/item?id=48841384</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=48841384</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48841384</guid></item><item><title><![CDATA[New comment by MasterScrat in "MIRA: Multiplayer Interactive World Models Trained on Rocket League"]]></title><description><![CDATA[
<p>Hey all, happy to see this here! This was a colab between General Intuition (that I’m part of), Kyutai and Epic Games.<p>You can read plenty of details in the blog post and tech report but the TLDR is that we trained a multiplayer world model on 10k hours of Rocket League data. We optimized it to be playable at 20fps on a single GPU.<p>So what you see in the demo is fully generated: there’s no graphics or physics engine. Instead it’s a 5b neural network that takes actions in and gives pixels out.</p>
]]></description><pubDate>Thu, 09 Jul 2026 01:47:47 +0000</pubDate><link>https://news.ycombinator.com/item?id=48839923</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=48839923</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48839923</guid></item><item><title><![CDATA[New comment by MasterScrat in "MIRA: Multiplayer Interactive World Models Trained on Rocket League"]]></title><description><![CDATA[
<p>We're happy to release MIRA, a collaboration between General Intuition, Kyutai, and Epic Games.<p>Mira was trained on 10k hours of Rocket League data. The model has 5B parameters and runs 4-player games at 20 fps on a single B200 GPU.<p>We've released a playable online demo, an in-depth technical report as well as a 1k hour dataset of 4-players gameplay:<p>Technical report: <a href="https://mira-wm.com/paper" rel="nofollow">https://mira-wm.com/paper</a>
Repo: <a href="https://github.com/mira-wm/mira" rel="nofollow">https://github.com/mira-wm/mira</a></p>
]]></description><pubDate>Tue, 07 Jul 2026 15:12:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=48818950</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=48818950</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48818950</guid></item><item><title><![CDATA[New comment by MasterScrat in "Show HN: MIRA – Multiplayer World Model Trained on Rocket League"]]></title><description><![CDATA[
<p>The checkpoint also weights half less than the game install! ;-)</p>
]]></description><pubDate>Tue, 07 Jul 2026 15:05:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=48818865</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=48818865</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48818865</guid></item><item><title><![CDATA[Show HN: MIRA – Multiplayer World Model Trained on Rocket League]]></title><description><![CDATA[
<p>Article URL: <a href="https://mira-wm.com/">https://mira-wm.com/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=48818661">https://news.ycombinator.com/item?id=48818661</a></p>
<p>Points: 12</p>
<p># Comments: 1</p>
]]></description><pubDate>Tue, 07 Jul 2026 14:51:27 +0000</pubDate><link>https://mira-wm.com/</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=48818661</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48818661</guid></item><item><title><![CDATA[Meta's AI Storage Blueprint at Scale]]></title><description><![CDATA[
<p>Article URL: <a href="https://engineering.fb.com/2026/07/01/data-infrastructure/metas-ai-storage-blueprint-at-scale/">https://engineering.fb.com/2026/07/01/data-infrastructure/metas-ai-storage-blueprint-at-scale/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=48800791">https://news.ycombinator.com/item?id=48800791</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Mon, 06 Jul 2026 04:57:50 +0000</pubDate><link>https://engineering.fb.com/2026/07/01/data-infrastructure/metas-ai-storage-blueprint-at-scale/</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=48800791</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48800791</guid></item><item><title><![CDATA[New comment by MasterScrat in "Task Failed Successfully: Saturating NIC and Disk Bandwidth"]]></title><description><![CDATA[
<p>This sounds like a strong statement with little backing. The author does infra at DeepSeek if his LinkedIn is to be trusted, and is the author of Foyer.</p>
]]></description><pubDate>Sat, 27 Jun 2026 15:06:08 +0000</pubDate><link>https://news.ycombinator.com/item?id=48698915</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=48698915</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48698915</guid></item><item><title><![CDATA[Machines do not press play: From temporal compression to model-ready tensors]]></title><description><![CDATA[
<p>Article URL: <a href="https://spiraldb.com/deep-dives/video">https://spiraldb.com/deep-dives/video</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=48576317">https://news.ycombinator.com/item?id=48576317</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Wed, 17 Jun 2026 20:23:28 +0000</pubDate><link>https://spiraldb.com/deep-dives/video</link><dc:creator>MasterScrat</dc:creator><comments>https://news.ycombinator.com/item?id=48576317</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48576317</guid></item></channel></rss>