<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>AI on Dmitry Bryndin</title><link>https://bryndin.com/categories/ai/</link><description>Recent content in AI on Dmitry Bryndin</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Fri, 05 Dec 2025 00:59:39 -0800</lastBuildDate><atom:link href="https://bryndin.com/categories/ai/index.xml" rel="self" type="application/rss+xml"/><item><title>Will Search Quality Determine the Next Generation of LLMs?</title><link>https://bryndin.com/posts/2025-12-05-will-search-quality-determine-the-next-generation-of-llms/</link><pubDate>Fri, 05 Dec 2025 00:59:39 -0800</pubDate><guid>https://bryndin.com/posts/2025-12-05-will-search-quality-determine-the-next-generation-of-llms/</guid><description>Advanced LLM modes like &amp;ldquo;research&amp;rdquo; and &amp;ldquo;thinking&amp;rdquo; depend heavily on web search, but they’re bounded by cost and scale. The quality of the search engine directly shapes the quality of the model’s output, giving LLM developers with strong search partners a clear edge in both accuracy and efficiency. Let’s take a closer look at how this all works.</description></item><item><title>Math Problems to Troll Your Students and LLMs</title><link>https://bryndin.com/posts/2025-03-17-math-problems-to-troll-your-students-and-llms/</link><pubDate>Mon, 17 Mar 2025 14:51:06 -0700</pubDate><guid>https://bryndin.com/posts/2025-03-17-math-problems-to-troll-your-students-and-llms/</guid><description>Designing deceptively simple math problems to challenge (or troll) students and LLMs, and why it is crucial to include enough atypical variations in training datasets and textbooks.</description></item><item><title>Comparing LLMs for Writing</title><link>https://bryndin.com/posts/2024-07-02-llms-for-writing/</link><pubDate>Tue, 02 Jul 2024 14:32:35 -0700</pubDate><guid>https://bryndin.com/posts/2024-07-02-llms-for-writing/</guid><description>Here, we explore how various LLMs perform in writing tasks, examining each one&amp;rsquo;s unique style and providing recommendations based on their strengths. Additionally, we delve into why Google&amp;rsquo;s Gemini is a misfit for writing, discussing the reasons behind its peculiar behavior.</description></item><item><title>Will we need to wait 10 years for the next wave in innovation?</title><link>https://bryndin.com/posts/2024-06-29-tech-waves/</link><pubDate>Sat, 29 Jun 2024 15:54:20 -0700</pubDate><guid>https://bryndin.com/posts/2024-06-29-tech-waves/</guid><description>Every 10 years the same comic strip comes up about the new technology: SQL database, blockchain, and now AI. When will be the next wave and what will it be about?&amp;hellip;</description></item></channel></rss>