<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Applied_ai on Bogdan Buduroiu</title><link>https://buduroiu.com/topics/applied_ai/</link><description>Recent content in Applied_ai on Bogdan Buduroiu</description><generator>Hugo</generator><language>en</language><copyright>All text licensed is licensed under CC BY-NC-SA 4.0 License.</copyright><lastBuildDate>Tue, 25 Aug 2026 01:34:50 +0300</lastBuildDate><atom:link href="https://buduroiu.com/topics/applied_ai/index.xml" rel="self" type="application/rss+xml"/><item><title>Speculative Programmatic Tool Calling</title><link>https://buduroiu.com/links/speculative-programmatic-tool-calling/</link><pubDate>Tue, 25 Aug 2026 01:34:50 +0300</pubDate><guid>https://buduroiu.com/links/speculative-programmatic-tool-calling/</guid><description>&lt;p&gt;This feels like a natural evolution of the RLM paradigm. With most interpreted languages, you don&amp;rsquo;t need to wait for the entire code block to be generated for you to start executing calls.&lt;/p&gt;</description></item></channel></rss>