<?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 Goran Mekić</title><link>https://meka.rs/tags/ai/</link><description>Recent content in Ai on Goran Mekić</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Wed, 11 Nov 2026 02:36:00 +0000</lastBuildDate><atom:link href="https://meka.rs/tags/ai/index.xml" rel="self" type="application/rss+xml"/><item><title>Opinion on AI and the future of IT</title><link>https://meka.rs/blog/2026/11/11/opinion-on-ai-and-the-future-of-it/</link><pubDate>Wed, 11 Nov 2026 02:36:00 +0000</pubDate><guid>https://meka.rs/blog/2026/11/11/opinion-on-ai-and-the-future-of-it/</guid><description>&lt;h3 id="current-state"&gt;&#10; Current State&#10;&lt;/h3&gt;&#10;&lt;p&gt;When a statement starts with &amp;ldquo;people often ask me &amp;hellip;&amp;rdquo;, you know it never happened, but during the&#10;past 9 months of job interviews, I was asked &amp;ldquo;What are your thoughts on AI and agentic&#10;development?&amp;rdquo; on a regular basis. Once I found a job, I forgot about the question until I met a&#10;friend at a store who asked me pretty much the same thing. My answer has always been &amp;ldquo;It&amp;rsquo;s like&#10;having a really fast bike, car, or Formula One&amp;rdquo;. What it means is that it depends on the driver:&#10;if it&amp;rsquo;s a lousy driver, the crash is going to happen pretty fast, but if the driver is good,&#10;everything will work perfectly.&lt;/p&gt;&#10;&lt;p&gt;That being said, I consider myself a pretty good AI driver. For the past year or so, I have worked&#10;on a Digital Audio Workstation, or DAW for short, in Rust. At first, it was a way to practice what&#10;I learned from the Rust book, but then it became a serious project. After reading that book, I&#10;needed a problem to solve in order to really understand the language and libraries, so I started&#10;rewriting &lt;a href="https://maolan.rs"&gt;Maolan&lt;/a&gt;. I started writing Maolan in January of 2018 in C++, and as I&#10;really liked the Rust language, I decided to give the rewrite a try. It went really well, and after&#10;two months of writing code, I had an audio backend, an engine, and the foundation in place. I have&#10;to admit, I was inspired by my ex-colleague, Alan Somers, who really likes Rust and whose opinion&#10;and knowledge I highly respect.&lt;/p&gt;&#10;&lt;p&gt;After writing the most important parts by hand, I decided to give AI a try. So far, I have tried&#10;OpenAI, Claude, Gemini, and Kimi. Gemini is not something I consider a good AI agent, but the other&#10;three are great. To be honest, Kimi costs the least, so I have used it the most. AI writes code&#10;insanely fast, and most of the time I have some nitpicks about the code it wrote; sometimes I just&#10;say &amp;ldquo;forget it, I&amp;rsquo;ll do this manually&amp;rdquo;. The reason is that AI doesn&amp;rsquo;t really know what you want,&#10;and if you&amp;rsquo;re not precise, code gets bad pretty fast. As you can see, I keep saying &amp;ldquo;fast&amp;rdquo;, and&#10;that&amp;rsquo;s why I think the fast vehicle analogy is good. In any case, Maolan is at the point where I&#10;started recording the first song using it, which revealed a bunch of bugs that I fixed.&lt;/p&gt;&#10;&lt;p&gt;My background is that I started playing guitar when I was 6, producing trance music when I was 16,&#10;working with electronics and schematics since I was 16, and programming since I was 19. Now, in my&#10;mid-forties, I have a lot of experience that makes me an expert in the field of audio/MIDI DSP and&#10;music production. I know exactly what I want and how to get it, and I know exactly what the&#10;hardware and kernel are capable of and how to use them. With such experience, AI is a huge boost,&#10;and when it goes sideways, I can tell right away.&lt;/p&gt;&#10;&lt;p&gt;As an experiment, I worked on a project in a domain I knew nothing about. I wanted to see how badly&#10;it would turn out, even with me as a senior developer and a pretty good IT architect. Within two&#10;months, I stopped, learned about the technologies I knew nothing about, read the AI-generated code,&#10;and threw it all away. That was me being a bad driver to prove my point to myself, as I hate having&#10;an opinion about something I&amp;rsquo;ve never tested.&lt;/p&gt;&#10;&lt;p&gt;I think I found my way of working with and learning from AI. Find your speed and don&amp;rsquo;t drive above&#10;that limit. Don&amp;rsquo;t use AI only to develop something; ask it a bunch of questions. I think a quarter&#10;of the tokens I spent in the first 6 months were for coding, and the other three-quarters went to&#10;AI explaining things to me and telling me what to read. Use AI as your teacher, as it can improve&#10;your coding skills a lot. And I mean A LOT!!!&lt;/p&gt;&#10;&lt;p&gt;As for programming languages, I think stricter ones are going to be more popular. My experience&#10;with AI and Rust is that Rust&amp;rsquo;s strictness makes AI do a better job. For Rust itself, when you add&#10;tools like rustfmt and clippy, along with tests, the results are even better. Tools like Miri can&#10;additionally guide AI models on how to write better code. Today, most of the new AI-written&#10;applications I see are written in Rust, and I think the strictness has a lot to do with it. If some&#10;future programming language becomes even stricter, I think it will be even better for AI-assisted&#10;development. Maybe some future tool for an existing language will catch bugs by analyzing the code&#10;even better; maybe it doesn&amp;rsquo;t have to be a new language. In any case, the more help AI gets, the&#10;better the code it will write. And not just AI, humans, too. When I started with Rust, I hated it&#10;because it prevented me from writing code the way I was used to, only to later realize why it does&#10;that and what the benefits of writing code the Rust way are.&lt;/p&gt;&#10;&lt;h3 id="the-future"&gt;&#10; The Future&#10;&lt;/h3&gt;&#10;&lt;p&gt;The logical question is &amp;ldquo;what&amp;rsquo;s next?&amp;rdquo;, so this is my prediction. Looking at the history of IT,&#10;everything starts centralized, then big companies have it in-house, then smaller companies have it&#10;in-house, then everyone has it. From ENIAC to home PCs, mobile phones, and tablets, it&amp;rsquo;s always&#10;pretty much the same. That being said, I think something similar will happen with AI; the only&#10;question is how exactly. I think the first step will be the widespread adoption of systems with&#10;large pools of unified memory, allowing today&amp;rsquo;s huge models to run on tomorrow&amp;rsquo;s personal machines.&#10;Of course, at first those systems will be more expensive than a gaming PC, so only companies will&#10;be able to afford them, but over time they will reach the masses, and it will be normal for your AI&#10;model to use most of your RAM during development.&lt;/p&gt;&#10;&lt;p&gt;At some point, I think we are bound to switch to optical computers, for two main reasons: power and&#10;speed. When I say power, I mean consumption of electricity. Moving electrons through resistive&#10;materials consumes energy and generates heat; photons can avoid those thermal losses. As CPU clock&#10;speeds have hovered around 5 GHz for quite a few years, much of the performance scaling has come&#10;from adding more cores. For processes that can be parallelized, that&amp;rsquo;s great, but sometimes you&#10;can&amp;rsquo;t do that. For example, in a single chain of audio effects, each effect must be processed in&#10;order, so multiple cores do not speed up that chain. Optical computers could potentially reach much&#10;higher frequencies. The problem today is that we still lack practical, scalable optical switching&#10;components (or optical transistors, if you will) that can replace their electronic counterparts in&#10;general-purpose computers. Transistors are fundamental building blocks of digital electronics, so&#10;we need an equivalent building block for digital optics. If you ever see news that such optical&#10;switching circuits have become practical and manufacturable at scale, that&amp;rsquo;s when we will gradually&#10;switch to optical computers. I can imagine a 1 kW optical computer as powerful as today&amp;rsquo;s small&#10;data center and requiring almost no cooling. I say &amp;ldquo;almost&amp;rdquo; because UPSes will still generate heat.&lt;/p&gt;&#10;&lt;p&gt;For the more distant future, I think a combination of optical and quantum is the way to go. In&#10;essence, the optical portion will do what electronic hardware does today with almost no need for&#10;programming to change, while quantum processors will solve specific classes of problems where they&#10;offer an advantage. I say &amp;ldquo;almost no need for change&amp;rdquo; because firmware will change drastically, but&#10;the kernel and userspace will not change as much. To use a quantum processor, or maybe a&#10;co-processor, programming will have to change drastically, but for an optical processor, it may not&#10;change as much.&lt;/p&gt;&#10;</description></item></channel></rss>