<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>l0l1 — Skelf Research</title><description>Notes on LLM-assisted SQL: validation, PII detection, and what privacy means when your copilot can see your warehouse.</description><link>https://l0l1.skelfresearch.com/</link><language>en-us</language><item><title>What &apos;privacy&apos; means when your copilot sees your warehouse</title><link>https://l0l1.skelfresearch.com/blog/privacy-when-copilot-sees-warehouse/</link><guid isPermaLink="true">https://l0l1.skelfresearch.com/blog/privacy-when-copilot-sees-warehouse/</guid><description>Privacy in AI data tools is usually marketing language. Here is a more honest account of what the threat model looks like, what l0l1 actually does, and what it explicitly doesn&apos;t claim.</description><pubDate>Fri, 22 May 2026 00:00:00 GMT</pubDate></item><item><title>Validation as a first-class layer in AI data tools</title><link>https://l0l1.skelfresearch.com/blog/validation-first-class-layer/</link><guid isPermaLink="true">https://l0l1.skelfresearch.com/blog/validation-first-class-layer/</guid><description>If LLMs are going to draft SQL, validation has to stop being a UI warning and start being an architectural surface. Here is how we think about it inside l0l1.</description><pubDate>Fri, 01 May 2026 00:00:00 GMT</pubDate></item><item><title>LLM-generated SQL is mostly correct — that&apos;s the problem</title><link>https://l0l1.skelfresearch.com/blog/llm-sql-mostly-correct/</link><guid isPermaLink="true">https://l0l1.skelfresearch.com/blog/llm-sql-mostly-correct/</guid><description>Why &apos;looks right and runs&apos; is the most dangerous failure mode for AI-generated SQL, and what a validation layer needs to catch before the dashboard updates.</description><pubDate>Sun, 12 Apr 2026 00:00:00 GMT</pubDate></item></channel></rss>