<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Lookahead-Bias on zolty.systems</title><link>https://blog.zolty.systems/tags/lookahead-bias/</link><description>Recent content in Lookahead-Bias on zolty.systems</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Mon, 04 May 2026 20:00:00 -0400</lastBuildDate><atom:link href="https://blog.zolty.systems/tags/lookahead-bias/index.xml" rel="self" type="application/rss+xml"/><item><title>Blind Oracle: stripping dates, headlines, and tickers before trusting an LLM trading evaluator</title><link>https://blog.zolty.systems/posts/2026-05-04-blind-oracle-llm-trading-evaluator/</link><pubDate>Mon, 04 May 2026 20:00:00 -0400</pubDate><guid>https://blog.zolty.systems/posts/2026-05-04-blind-oracle-llm-trading-evaluator/</guid><description>How I caught my LLM-driven hypothesis evaluator cheating by reading the future — and the masking patch that finally produced trustworthy out-of-sample numbers.</description></item></channel></rss>