You Had Better Catch Up. Fast.
Next, exclusively on Substack: We are going inside the new generation of compilers to show what they can actually do, how much more of a program’s meaning they can understand, and how they can help fix the mistakes today’s AI coders still make.
The interesting part is what happens when you put both technologies together. AI coders still have to guess a lot of things. Compilers with explicit semantics can take much of that guesswork out of the process.
The result could be a way to turn semi-formal prompts into safe, reliable code, and even work backwards to decompile complex software and figure out what is going on under the hood.
For now, here are two animations to give you a taste of what is coming, with several more in the full article.
Both examples focus on the latest advances in C++ compilers, but there is a lot more to cover. And of course, we will go well beyond C++.
Let us start with two basic animations to give you a glimpse of what is coming in the full article, arriving in just a few days
Sets, mappings, and nested contexts. How context narrows the meaning of words and concepts, and how that meaning can be carried safely from an entire project down to a single C++ function.
How XS compilers are quietly taking on AI coders. A closer look at GCC 16.1 and C++26, and the new mechanisms that allow compilers to understand things AI coders still struggle with.
Want to go deeper and learn some seriously useful techniques you simply will not find anywhere else? This is just the beginning. Dozens more articles are on the way, packed with practical examples, real-world applications, and fresh insights into some of the most important areas of technology and AI.
If you are NOT a paid subscriber yet, check out the note below.
And if you already are, congratulations! The full article will land in your inbox in just a few days, exclusively for paid subscribers.
A Note for Free Subscribers ONLY:
The full version of this article is for paid subscribers only. We are going hands-on with XS compilers and AI coders, showing how to put them to work together to turn semi-formal prompts into reliable code and decompile complex software.
Less hype, more actual engineering. Practical examples, real code, and a look at what these new compilers can do that AI coders still cannot.
Coming in the next few days. Upgrade now and get the full article as soon as it drops.