We have a service at work which categorizes internal documents and logs, then triggers some automation depending on the category. It processes maybe 100 per day. Previously we only used some combination of metadata, regex, and NLP to categorize. Now a call to a LLM is part of that service. We save a lot of manual time where we used to have to resolve unknown documents. The LLM can help fill out missing data, too. It's all stored as annotations so it's clear who/what edited the data.
Granted this is a pretty simple task and a low stakes scenario, but I don't think we should limit ourselves to assuming AI will always only be dev tooling.
I used Btrfs for a few years but switched away a couple years ago. I also had one or two incidents with Btrfs where some weirdness happened, but I was able to recover everything in the end. Overall I liked the flexibility of Btrfs, but mostly I found it too slow.
I use ZFS on Arch Linux and overall have had no problems with it so far. There's more customization and methods to optimize performance. My one suggestion is to do a lot of research and testing with ZFS. There is a bit of a learning curve, but it's been worth the switch for me.
Kudos for one of the most relatable descriptions of C++ I've read.
I did a couple years writing C/C++ professionally, and I hope to not go back to that. Too many hours debugging other people's code, suffering vague integration issues, and just trying to get the build system spaghetti to run.
I have fond memories of reading this book as a teenager. I was watching an introduction to computer science course online. I think maybe from Yale or MIT, and the professor recommended Hacker's Delight. It was a beginner class teaching C, and I mostly only had experience with BASIC.
Back then I barely understood binary, and pointers completely confused me. I remember most of the book feeling like a collection of magic tricks. Sometimes I pull it out to rekindle that sense of wonder.
I think it is completely worth it to learn the Vim editing commands. You can get a lot of the benefits from just turning on Vim-mode in VSCode or IntelliJ. Emacs with Evil mode is an improvement in my opinion as well. The quick line editing, moving around the file, etc etc smooth out your programming experience a lot.
If you're working a lot with text, Vim macros are great. I'll regularly go into Vim as kind of a text workbench.
If you want to try an auto-updating Vim suite, check out LazyVim [0]. The defaults are great, and there's a lot of features with absolutely zero configuration.
I was thinking the same thing. A 90 minute movie where he doesn't say a single word. It could be one day in his life just being there for his different clients.
Granted this is a pretty simple task and a low stakes scenario, but I don't think we should limit ourselves to assuming AI will always only be dev tooling.