<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
<channel>
<title>pawan</title>
<link>https://cintu07.github.io/</link>
<description>long posts on ml infra, databases and inference, and the bugs in between.</description>
<atom:link href="https://cintu07.github.io/feed.xml" rel="self" type="application/rss+xml"/>
<item>
<title>gradient descent is one line of code. here is everything that line hides.</title>
<link>https://cintu07.github.io/posts/gradient-descent/</link>
<guid>https://cintu07.github.io/posts/gradient-descent/</guid>
<pubDate>Sat, 03 Oct 2026 00:00:00 +0000</pubDate>
<description>the line almost every neural network is trained with, built from nothing and measured: when it is fast, slow, or explodes, and what sgd, momentum, adam and backprop each change.</description>
</item>
<item>
<title>your ai has amnesia, and a bigger context window will not fix it</title>
<link>https://cintu07.github.io/posts/your-ai-has-amnesia/</link>
<guid>https://cintu07.github.io/posts/your-ai-has-amnesia/</guid>
<pubDate>Mon, 10 Aug 2026 00:00:00 +0000</pubDate>
<description>three turns after a user removed their card, i watched the agent tell them it was still on file. same session.</description>
</item>
<item>
<title>ciot: a cpu inference engine for ternary networks</title>
<link>https://cintu07.github.io/posts/ciot-ternary-inference/</link>
<guid>https://cintu07.github.io/posts/ciot-ternary-inference/</guid>
<pubDate>Tue, 26 May 2026 00:00:00 +0000</pubDate>
<description>0.54 ms per 1024x1024 ternary matvec on a snapdragon x laptop. zero dependencies, one binary, ~3,000 lines of c++ and python.</description>
</item>
<item>
<title>Building a Membership System That Says &quot;I Don&#x27;t Know&quot;</title>
<link>https://cintu07.github.io/posts/styx-membership-system/</link>
<guid>https://cintu07.github.io/posts/styx-membership-system/</guid>
<pubDate>Fri, 24 Oct 2025 00:00:00 +0000</pubDate>
<description>Most failure detectors lie. STYX is a distributed membership system that returns a probability distribution instead of a boolean, handling uncertainty with honesty.</description>
</item>
</channel>
</rss>
