<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Devin de Hueck</title>
    <description>The latest articles on DEV Community by Devin de Hueck (@ddehueck).</description>
    <link>https://dev.to/ddehueck</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F563156%2F23a1ca6d-6d31-4f58-b097-3262c1a61358.gif</url>
      <title>DEV Community: Devin de Hueck</title>
      <link>https://dev.to/ddehueck</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/ddehueck"/>
    <language>en</language>
    <item>
      <title>I made an infinite canvas to work with LLMs better</title>
      <dc:creator>Devin de Hueck</dc:creator>
      <pubDate>Sat, 20 Apr 2024 13:20:28 +0000</pubDate>
      <link>https://dev.to/ddehueck/i-made-an-infinite-canvas-to-work-with-llms-better-dgj</link>
      <guid>https://dev.to/ddehueck/i-made-an-infinite-canvas-to-work-with-llms-better-dgj</guid>
      <description>&lt;p&gt;&lt;a href="https://media.dev.to/cdn-cgi/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fj2fwryk5s56km29lm54r.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media.dev.to/cdn-cgi/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fj2fwryk5s56km29lm54r.png" alt="Image description" width="800" height="500"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;This project came about after talking with a friend who was trying to use ChatGPT to help him write a research paper by doing simple things like incorporating notes, rephrasing sections, building outlines and sub-outlines.&lt;/p&gt;

&lt;p&gt;However, he described this process as painful as he needed to constantly be switching windows and tabs for each of these subtasks - each of which required different context (notes that would be copy/pasted into the chat).&lt;/p&gt;

&lt;p&gt;I've become convinced that a chat interface is not the right way to work with AI when the process needs to be collaborative and iterative. Which is why I'm building: &lt;a href="https://lxi.ai"&gt;https://lxi.ai&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;You can do things like:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Store AI generated text/outlines you like on the canvas that you can then edit directly.&lt;/li&gt;
&lt;li&gt;Control the context when using chatting or generating outputs with AI (Just connect the dots)&lt;/li&gt;
&lt;li&gt;Apply the same prompt over multiple inputs (Just connect the dots)&lt;/li&gt;
&lt;li&gt;Share the canvas with a teammate and collaborate on the same canvas in real-time&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Some use cases:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Draft that blog post you've been thinking about (hint: try starting with an outline)&lt;/li&gt;
&lt;li&gt;Plan that trip you've been wanting to take (hint: create your daily itinerary then use the function node to create an hourly itinerary)&lt;/li&gt;
&lt;li&gt;Draft some social media posts (hint: Add your landing page copy to a text box, connect to a AI Generator node)&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Overall, I'm not into this whole replace or a one and done approach to using AI when making things. I'd much rather like to see AI accelerate the tedious parts which is what I'm hoping lxi will do.&lt;/p&gt;

&lt;p&gt;Thanks for checking it out and please create a free account and let me know if you have any feedback in the comments!&lt;/p&gt;

&lt;p&gt;Link: &lt;a href="https://lxi.ai"&gt;https://lxi.ai&lt;/a&gt;&lt;br&gt;
Built with: sveltekit, Liveblocks, lucia auth, neon&lt;/p&gt;

</description>
      <category>sideprojects</category>
      <category>svelte</category>
      <category>sveltekit</category>
      <category>ai</category>
    </item>
    <item>
      <title>I made a SMS app to text my ideas to and it'll text them back at random intervals in the future</title>
      <dc:creator>Devin de Hueck</dc:creator>
      <pubDate>Mon, 18 Jul 2022 01:26:23 +0000</pubDate>
      <link>https://dev.to/ddehueck/i-made-a-sms-app-to-text-my-ideas-to-and-itll-text-them-back-at-random-intervals-in-the-future-3lmh</link>
      <guid>https://dev.to/ddehueck/i-made-a-sms-app-to-text-my-ideas-to-and-itll-text-them-back-at-random-intervals-in-the-future-3lmh</guid>
      <description>&lt;p&gt;I'm someone who always texts myself my ideas as I think of them at random moments. I really enjoy the simplicity of texting ideas but I wanted a way to find my old ideas so I made &lt;a href="https://galaxi.me"&gt;https://galaxi.me&lt;/a&gt;. Let me know what you think!&lt;/p&gt;

&lt;p&gt;Just text an idea to: +17755225513&lt;/p&gt;

&lt;h2&gt;
  
  
  Tech Stack
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;FastAPI&lt;/li&gt;
&lt;li&gt;Postgres&lt;/li&gt;
&lt;li&gt;Procrastinate (PostgreSQL-based Task Queue)&lt;/li&gt;
&lt;li&gt;Twilio API&lt;/li&gt;
&lt;/ul&gt;

</description>
      <category>sideprojects</category>
      <category>twilio</category>
    </item>
  </channel>
</rss>
