<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Kevin Gibson's personal site</title><link>https://www.kevin-gibson.com/</link><description>Recent content on Kevin Gibson's personal site</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Sat, 02 May 2026 10:44:53 -0700</lastBuildDate><atom:link href="https://www.kevin-gibson.com/index.xml" rel="self" type="application/rss+xml"/><item><title>Writing a bindless GPU abstraction layer</title><link>https://www.kevin-gibson.com/blog/writing-a-bindless-gpu-abstraction-layer/</link><pubDate>Sat, 02 May 2026 10:44:53 -0700</pubDate><guid>https://www.kevin-gibson.com/blog/writing-a-bindless-gpu-abstraction-layer/</guid><description>&lt;figure&gt;&lt;a href="https://github.com/rkevingibson/loon_gpu"&gt;&lt;img src="https://www.kevin-gibson.com/img/loon_gpu.svg"
 alt="The Loon GPU library logo, a line drawing of the head of a common loon."&gt;&lt;/a&gt;
&lt;/figure&gt;

&lt;p&gt;Back in December 2025, Sebastian Aaltonen published a blog titled &lt;a href="https://www.sebastianaaltonen.com/blog/no-graphics-api"&gt;&amp;ldquo;No Graphics API&amp;rdquo;&lt;/a&gt; - it presented a great history of the evolution of GPU hardware, and gave an opinionated perspective on how we could simplify the modern graphics APIs on modern hardware. Like many graphics programmers, I read Sebastian&amp;rsquo;s blog post and really enjoyed it - I was inspired by it, and decided to see how close to the API he describes I could get today, layered on top of existing platform APIs. The answer turns out to be &amp;ldquo;pretty darn close&amp;rdquo;. The result is a project I&amp;rsquo;m calling Loon GPU, and I&amp;rsquo;ve put it up on &lt;a href="https://github.com/rkevingibson/loon_gpu"&gt;Github&lt;/a&gt;. While the library is still rough, poorly tested and surely filled with bugs, it is usable and I wanted to share it early in case other folks were interested. Currently it has a Vulkan 1.3 and Metal 4 backend, and here I want to do a high-level walkthrough to see how the API design maps on to those backends.&lt;/p&gt;</description></item><item><title>Neural networks as Ordinary Differential Equations</title><link>https://www.kevin-gibson.com/blog/neural-networks-as-ordinary-differential-equations/</link><pubDate>Tue, 11 Dec 2018 18:46:36 -0800</pubDate><guid>https://www.kevin-gibson.com/blog/neural-networks-as-ordinary-differential-equations/</guid><description>&lt;p&gt;Recently I found a paper being presented at NeurIPS this year, entitled &lt;a href="https://arxiv.org/abs/1806.07366"&gt;Neural Ordinary Differential Equations&lt;/a&gt;, written by Ricky Chen, Yulia Rubanova, Jesse Bettencourt, and David Duvenaud from the University of Toronto. The core idea is that certain types of neural networks are analogous to a discretized differential equation, so maybe using off-the-shelf differential equation solvers will help get better results. This led me down a bit of a rabbit hole of papers that I found very interesting, so I thought I would share a short summary/view-from-30,000 feet on this idea.&lt;/p&gt;</description></item><item><title>About</title><link>https://www.kevin-gibson.com/about/</link><pubDate>Sun, 18 Nov 2018 16:32:06 -0800</pubDate><guid>https://www.kevin-gibson.com/about/</guid><description>&lt;p&gt;I am a software engineer with experience in computer graphics, computer vision, and low level systems. Most recently I&amp;rsquo;ve worked at Apple, developing a 3D Gaussian Splat renderer that is used by the &lt;a href="https://www.cnet.com/tech/computing/apple-talks-to-me-about-vision-pro-personas-where-is-our-virtual-presence-headed/"&gt;Vision Pro&amp;rsquo;s Personas&lt;/a&gt;. Previously I was at Microsoft, working on streaming systems for Xbox cloud gaming and surface reconstruction algorithms for Hololens, along with a brief stint working on firmware for Xbox controllers.&lt;/p&gt;</description></item></channel></rss>