<?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 Antoine Boucher</title><link>https://antoineboucher.info/CV/blog/tags/ai/</link><description>Recent content in AI on Antoine Boucher</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Mon, 10 Apr 2023 10:00:00 -0400</lastBuildDate><atom:link href="https://antoineboucher.info/CV/blog/tags/ai/index.xml" rel="self" type="application/rss+xml"/><item><title>GPT-4 vs GPT-3.5 — capabilities and API cost framing</title><link>https://antoineboucher.info/CV/blog/posts/gpt4-api-costs-overview/</link><pubDate>Mon, 10 Apr 2023 10:00:00 -0400</pubDate><guid>https://antoineboucher.info/CV/blog/posts/gpt4-api-costs-overview/</guid><description>&lt;p&gt;&lt;img src="https://media.licdn.com/dms/image/D5612AQHKus7rY_HvVQ/article-cover_image-shrink_720_1280/0/1681059889367?e=1709769600&amp;amp;v=beta&amp;amp;t=39LUP6caikwBW_QeEC7T-2peoci56x6v9xzNtCPdWxQ" alt="Article cover image" title="Cover Image for AI Article"&gt;&lt;/p&gt;
&lt;p&gt;As artificial intelligence continues to advance, more and more companies are incorporating AI-powered chatbots into their customer service systems. These chatbots can handle a wide range of customer inquiries, from simple questions to more complex issues. However, the cost of implementing and maintaining such chatbots is an important factor to consider. In this article, we will calculate the costs of using GPT-4 and GPT-3.5-turbo models with a message cap of 25 messages every 3 hours for one month of usage, considering the same average prompt sizes (50 to 200 tokens).&lt;/p&gt;</description></item><item><title>GitHub Copilot session at Cédille (with GitHub &amp; Arctiq)</title><link>https://antoineboucher.info/CV/blog/posts/github-copilot-cedille-session/</link><pubDate>Tue, 06 Sep 2022 10:00:00 -0400</pubDate><guid>https://antoineboucher.info/CV/blog/posts/github-copilot-cedille-session/</guid><description>&lt;p&gt;At &lt;strong&gt;Cédille&lt;/strong&gt;, we hosted a session with &lt;strong&gt;GitHub&lt;/strong&gt; and &lt;strong&gt;Arctiq&lt;/strong&gt; focused on &lt;strong&gt;GitHub Copilot&lt;/strong&gt; and AI-assisted development. Highlights included &lt;strong&gt;Copilot Chat&lt;/strong&gt; with &lt;code&gt;/createNotebook&lt;/code&gt; for quick Jupyter notebooks from existing code, and pointers to &lt;strong&gt;&lt;a href="https://githubnext.com"&gt;GitHub Next&lt;/a&gt;&lt;/strong&gt; experiments.&lt;/p&gt;
&lt;p&gt;Thanks to speakers &lt;strong&gt;Thierry Madkaud&lt;/strong&gt; and &lt;strong&gt;Eldrick Wega&lt;/strong&gt;, and to everyone who joined.&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;&lt;a href="https://antoineboucher.info/CV/blog/posts/github-copilot-cedille-session/"&gt;Full article in French&lt;/a&gt;&lt;/strong&gt; (same slug — you can also switch to &lt;strong&gt;FR&lt;/strong&gt; in the header).&lt;/p&gt;</description></item></channel></rss>