<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Nick Pizzo</title><description>Production-minded notes on LLM systems, evals, agent workflows, and the tradeoffs behind applied AI engineering.</description><link>https://nickpizzo.pizza/</link><item><title>Why Weaviate Query Agents make sense for shopping search</title><link>https://nickpizzo.pizza/posts/why-weaviate-agents-make-sense-for-shopping-search/</link><guid isPermaLink="true">https://nickpizzo.pizza/posts/why-weaviate-agents-make-sense-for-shopping-search/</guid><description>Weaviate&apos;s Query Agent pattern fits ecommerce search when semantic intent, structured filters, source traces, and latency budgets are treated as one system.</description><pubDate>Mon, 11 May 2026 14:16:42 GMT</pubDate></item><item><title>RAG-ready full-text search on Azure AI Search</title><link>https://nickpizzo.pizza/posts/standard-search-first-rag-ready-azure-ai-search/</link><guid isPermaLink="true">https://nickpizzo.pizza/posts/standard-search-first-rag-ready-azure-ai-search/</guid><description>I recommended Azure AI Search for predictable Contentful-backed search first, with the same retrieval foundation ready for a later RAG answer layer.</description><pubDate>Fri, 08 May 2026 20:33:27 GMT</pubDate></item></channel></rss>