<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: armanified</title><link>https://news.ycombinator.com/user?id=armanified</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Tue, 07 Apr 2026 00:06:44 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=armanified" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by armanified in "Show HN: I built a tiny LLM to demystify how language models work"]]></title><description><![CDATA[
<p>True, but most would ignore LM if it weren't LLM.</p>
]]></description><pubDate>Mon, 06 Apr 2026 12:47:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=47660234</link><dc:creator>armanified</dc:creator><comments>https://news.ycombinator.com/item?id=47660234</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47660234</guid></item><item><title><![CDATA[New comment by armanified in "Show HN: I built a tiny LLM to demystify how language models work"]]></title><description><![CDATA[
<p>This title might have triggered something in those bots; most of them have sneaky AI SaaS links in their bio.<p>Honestly, I never expected this post to become so popular. It was just the outcome of a weekend practice session.</p>
]]></description><pubDate>Mon, 06 Apr 2026 12:45:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=47660217</link><dc:creator>armanified</dc:creator><comments>https://news.ycombinator.com/item?id=47660217</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47660217</guid></item><item><title><![CDATA[New comment by armanified in "Show HN: I built a tiny LLM to demystify how language models work"]]></title><description><![CDATA[
<p>OMG! Why didn't I thought fo this first :P</p>
]]></description><pubDate>Mon, 06 Apr 2026 11:59:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=47659783</link><dc:creator>armanified</dc:creator><comments>https://news.ycombinator.com/item?id=47659783</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47659783</guid></item><item><title><![CDATA[New comment by armanified in "Show HN: I built a tiny LLM to demystify how language models work"]]></title><description><![CDATA[
<p>OMG! You just gave me the next  idea..</p>
]]></description><pubDate>Mon, 06 Apr 2026 11:57:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=47659761</link><dc:creator>armanified</dc:creator><comments>https://news.ycombinator.com/item?id=47659761</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47659761</guid></item><item><title><![CDATA[New comment by armanified in "Show HN: I built a tiny LLM to demystify how language models work"]]></title><description><![CDATA[
<p>Pretty neat! I'll definitely take a deeper look into this.</p>
]]></description><pubDate>Mon, 06 Apr 2026 11:53:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=47659724</link><dc:creator>armanified</dc:creator><comments>https://news.ycombinator.com/item?id=47659724</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47659724</guid></item><item><title><![CDATA[New comment by armanified in "Show HN: I built a tiny LLM to demystify how language models work"]]></title><description><![CDATA[
<p>Uppercase letters were intentionally ignored.</p>
]]></description><pubDate>Mon, 06 Apr 2026 11:51:48 +0000</pubDate><link>https://news.ycombinator.com/item?id=47659711</link><dc:creator>armanified</dc:creator><comments>https://news.ycombinator.com/item?id=47659711</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47659711</guid></item><item><title><![CDATA[New comment by armanified in "Show HN: I built a tiny LLM to demystify how language models work"]]></title><description><![CDATA[
<p>My initial idea was to train a navigation decision model with 25M parameters for a Raspberry Pi, which, in testing, was getting about 60% of tool calls correct. IMO, it seems like around 20M parameters would be a good size for following some narrow & basic language instructions.</p>
]]></description><pubDate>Mon, 06 Apr 2026 11:50:20 +0000</pubDate><link>https://news.ycombinator.com/item?id=47659698</link><dc:creator>armanified</dc:creator><comments>https://news.ycombinator.com/item?id=47659698</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47659698</guid></item><item><title><![CDATA[New comment by armanified in "Show HN: I built a tiny LLM to demystify how language models work"]]></title><description><![CDATA[
<p>It mostly doesn't, at 9M it has very limited capacity. The whole idea of this project is to demonstrate how Language Models work.</p>
]]></description><pubDate>Mon, 06 Apr 2026 11:41:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=47659630</link><dc:creator>armanified</dc:creator><comments>https://news.ycombinator.com/item?id=47659630</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47659630</guid></item><item><title><![CDATA[New comment by armanified in "Show HN: I built a tiny LLM to demystify how language models work"]]></title><description><![CDATA[
<p>I haven't compared it with anything yet. Thanks for the suggestion; I'll look into these.</p>
]]></description><pubDate>Mon, 06 Apr 2026 11:37:48 +0000</pubDate><link>https://news.ycombinator.com/item?id=47659601</link><dc:creator>armanified</dc:creator><comments>https://news.ycombinator.com/item?id=47659601</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47659601</guid></item><item><title><![CDATA[New comment by armanified in "Show HN: I built a tiny LLM to demystify how language models work"]]></title><description><![CDATA[
<p>I intentionally removed all optimizations to keep it vanilla.</p>
]]></description><pubDate>Mon, 06 Apr 2026 11:36:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=47659593</link><dc:creator>armanified</dc:creator><comments>https://news.ycombinator.com/item?id=47659593</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47659593</guid></item><item><title><![CDATA[Show HN: I built a tiny LLM to demystify how language models work]]></title><description><![CDATA[
<p>Built a ~9M param LLM from scratch to understand how they actually work. Vanilla transformer, 60K synthetic conversations, ~130 lines of PyTorch. Trains in 5 min on a free Colab T4. The fish thinks the meaning of life is food.<p>Fork it and swap the personality for your own character.</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47655408">https://news.ycombinator.com/item?id=47655408</a></p>
<p>Points: 837</p>
<p># Comments: 126</p>
]]></description><pubDate>Mon, 06 Apr 2026 00:20:12 +0000</pubDate><link>https://github.com/arman-bd/guppylm</link><dc:creator>armanified</dc:creator><comments>https://news.ycombinator.com/item?id=47655408</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47655408</guid></item><item><title><![CDATA[New comment by armanified in "Show HN: Online OCR Free – Batch OCR UI for Tesseract, Gemini and OpenRouter"]]></title><description><![CDATA[
<p>Good to see someone contributing to the Bangla space.</p>
]]></description><pubDate>Wed, 04 Mar 2026 13:29:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=47247126</link><dc:creator>armanified</dc:creator><comments>https://news.ycombinator.com/item?id=47247126</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47247126</guid></item><item><title><![CDATA[New comment by armanified in "Show HN: I built an HTTP client that perfectly mimics Chrome 142"]]></title><description><![CDATA[
<p>A lot was generated with LLM, but not vibe-coded. I have a clear expectation and project plan.</p>
]]></description><pubDate>Sun, 09 Nov 2025 20:08:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=45868691</link><dc:creator>armanified</dc:creator><comments>https://news.ycombinator.com/item?id=45868691</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45868691</guid></item><item><title><![CDATA[New comment by armanified in "Show HN: I built an HTTP client that perfectly mimics Chrome 142"]]></title><description><![CDATA[
<p>JS/TS could actually be my next venture! Would love to hear about your use case to help prioritize features if I build it out.</p>
]]></description><pubDate>Sun, 09 Nov 2025 12:37:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=45865117</link><dc:creator>armanified</dc:creator><comments>https://news.ycombinator.com/item?id=45865117</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45865117</guid></item><item><title><![CDATA[New comment by armanified in "Show HN: I built an HTTP client that perfectly mimics Chrome 142"]]></title><description><![CDATA[
<p>Fair concern. Most of these projects are abandoned because keeping up with major browsers' changes is exhausting. I genuinely want to maintain this and plan to update it with major Chrome releases as part of my work depends on it. Things move fast in this space, but I am committed to keeping it functional. I'll do my best to stay on top of it.</p>
]]></description><pubDate>Sun, 09 Nov 2025 12:36:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=45865110</link><dc:creator>armanified</dc:creator><comments>https://news.ycombinator.com/item?id=45865110</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45865110</guid></item><item><title><![CDATA[Show HN: I built an HTTP client that perfectly mimics Chrome 142]]></title><description><![CDATA[
<p>BoringSSL and nghttp2. Matches JA3N, JA4, and JA4_R fingerprints. Supports HTTP/2, async/await, and works with Cloudflare-protected sites. Not trying to compete with curl_cffi - just a learning project that turned into something functional.</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=45857312">https://news.ycombinator.com/item?id=45857312</a></p>
<p>Points: 39</p>
<p># Comments: 6</p>
]]></description><pubDate>Sat, 08 Nov 2025 15:28:24 +0000</pubDate><link>https://github.com/arman-bd/httpmorph</link><dc:creator>armanified</dc:creator><comments>https://news.ycombinator.com/item?id=45857312</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45857312</guid></item></channel></rss>