<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Stele on Yonk-Labs</title><link>https://yonk.dev/tags/stele/</link><description>Recent content in Stele on Yonk-Labs</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>© 2026 Yonk-Labs</copyright><lastBuildDate>Tue, 02 Jun 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://yonk.dev/tags/stele/index.xml" rel="self" type="application/rss+xml"/><item><title>Reading the Big-Ass Grid: A Field Guide to Our RAG Bake-Off</title><link>https://yonk.dev/blog/reading-the-grid/</link><pubDate>Tue, 02 Jun 2026 00:00:00 +0000</pubDate><guid>https://yonk.dev/blog/reading-the-grid/</guid><description>A 150-row benchmark grid looks like the output of a robot having a stroke — until you know the three things each row tells you. A field guide to reading our RAG bake-off: read the parametric floor first, decode the system and lane columns, and ask the only two questions that matter — is it right, and what did it cost?</description><media:content xmlns:media="http://search.yahoo.com/mrss/" url="https://yonk.dev/blog/reading-the-grid/feature.jpg"/></item><item><title>What Actually Moves RAG Accuracy (And What I Spent A Week Measuring Wrong)</title><link>https://yonk.dev/blog/what-actually-moves-rag-accuracy/</link><pubDate>Mon, 01 Jun 2026 00:00:00 +0000</pubDate><guid>https://yonk.dev/blog/what-actually-moves-rag-accuracy/</guid><description>One failing LoCoMo question turned into a cross-corpus, multi-system benchmark — and a pile of retracted conclusions. Small-N runs lie, cross-vendor numbers are rarely apples-to-apples, and a correctness bug will impersonate an architecture win every time. Run the no-context baseline, 6x your sample, and diff the bytes that reach the model before you trust any RAG number.</description><media:content xmlns:media="http://search.yahoo.com/mrss/" url="https://yonk.dev/blog/what-actually-moves-rag-accuracy/feature.jpg"/></item><item><title>Stele Tutorial: Five Minutes, Then Your Agent Remembers</title><link>https://yonk.dev/blog/stele-tutorial-five-minutes-then-your-agent-remembers/</link><pubDate>Sat, 23 May 2026 00:00:00 +0000</pubDate><guid>https://yonk.dev/blog/stele-tutorial-five-minutes-then-your-agent-remembers/</guid><description>The hands-on follow-up to the why-I-built-it post. Real commands, real outputs: install Stele, wire it into your agent, store artifacts with citations, supersede facts, time-travel with as_of, stash oversized tool output, and run recall through two strategies. Five minutes to install, the rest is just typing.</description><media:content xmlns:media="http://search.yahoo.com/mrss/" url="https://yonk.dev/blog/stele-tutorial-five-minutes-then-your-agent-remembers/feature.jpg"/></item><item><title>Stele: The Memory Layer I Couldn't Stop Building</title><link>https://yonk.dev/blog/stele-the-memory-layer-i-couldnt-stop-building/</link><pubDate>Thu, 21 May 2026 00:00:00 +0000</pubDate><guid>https://yonk.dev/blog/stele-the-memory-layer-i-couldnt-stop-building/</guid><description>I said the implementation needed another quarter. Three weeks later I&amp;rsquo;d shipped Stele — source-backed, time-traveling, sovereign agent memory that plugs into seven coding assistants. What it does, the three goals driving it, what&amp;rsquo;s solid on main, and what&amp;rsquo;s still wobbly. The honest version, including the parts that aren&amp;rsquo;t built yet.</description><media:content xmlns:media="http://search.yahoo.com/mrss/" url="https://yonk.dev/blog/stele-the-memory-layer-i-couldnt-stop-building/feature.jpg"/></item></channel></rss>