AI coding "assistants"

Solarion said:
Also many years coder here. I'm loving it. AI coding gives me spare time now to do other things and it takes the edge off.
Click to expand...
I've stepped away from coding, but miss it badly. All these tools taking away the leg work makes me think I might want to go back again
 
Built a new Node/Typescript project in about 2 days using Claude Code.

I had attempted basically the same thing a few months ago and it was significantly better this time around. It never got things horribly wrong like my first attempt where it would write code that just wouldn't run on quite a few occasions.

As Typescript isn't something I ever work with, it was definitely a massive time saver for something that we're just going to be using as an example template for other developers to get familiar with an API.

I basically started with writing documentation for the API and the schemas (mostly manually), wrote a few steps for a plan and I just guided it the rest of the way.

Having $40 of free credits was nice too!
 
Learning C# (windows desktop app) as a kind of a hobby. Coded in VBA for years. More as curiosity to see what I can do. Also, always satisfying when your code does things. Some aspects/concepts are complex and really hard. So many different ways to achieve an outcome. So many rules and best practices. Eventually, much can be grasped with experience but I cannot invest the time to learn and debug. Nor do I have a must achieve goal of any kind. So, I use AI and it churns out snippets of code, some quite extensive. I use, integrate and debug (with more help). Enough to do just to integrate into the half-baked code I started with. In 3 months I have a working something which could have taken me a year, but more likely I would have given up, and would have been full of bad practice/design, unrefactored, unreadable, buggy (not just edge cases) code.
 
Been busy with a little side project. Not sure it is that little anymore. I’ve just prompted and guided co-pilot, not written single line of code. Think the little project is about 30K+ lines of .py and .jsx
Dev ‘space’ is running across 6 containers, and think I’ll need a map/tile server as well.

It became a bit of a handful to vibecode, so I spun up a SonarQube container, and plugged that into the IDE and co-pilot. Seems like there is about 7% duplicate code, so I’ll have to hold back on features and do some refactoring.

It has become quite the undertaking to run on the side of a full time job and 11+ hours of cycling a week. Luckily the kid is grown up out the house and in own place, paying own bills.
 
Last edited:
InTheWild said:
Anthropic's leadership has been quite vocal on possible future outcomes in terms of jobs etc. I think the important part will be to keep up with all the models in terms of use and what it can achieve for you.

Anthropic also posted a new ad yesterday, where they do try to spin that you should still focus on solving problems, learning. But with AI:

https://twitter.com/x/status/1968705632095158393

In the end I think it's going to be important to identify when and how to use AI or not for certain things.

So I guess we should be getting in the mindset (I've been trying to get better with this) that we could be tackling any difficult problem now, because we have this massive new technology which is insane if I think how slow our projects used to be pre 2023.
Click to expand...
just watched this and realized at least 3 scenes in this ad was shot in Cape town.
 
Another big year for AI in the dev world. With 2025 drawing to a close, I thought I'd bump this thread to see where everyone's head is at as we start thinking about 2026.

I share so many of the concerns raised here it's all too easy to sit at the prompt like a mindless interpreter, churning out code that you no longer fully understand or even care to. Sure, you might do peer reviews, but those often boil down to syntax checks and minor logic tweaks.


Lately, I've been feeling like I need a complete overhaul of my architecture fundamentals. It seems like I'm forcing a modern tool into old-fashioned patterns, and it just doesn't feel natural anymore.

So, what's the big change you're planning for 2026? How are you going to reclaim the challenge and the joy of coding? I'd love to hear what everyone has in mind.
 
The biggest mindset change for me this year is that AI is better than me at coding (I've accepted it) but not better than me at project&task management overall. So I've shifted myself into a manager role, and managing AI as if its employees. I've managed multiple dev teams before, but moved into dev only the last few years. Ironic that AI is pushing me to a manager role again. This time though it's on another level.

I also think @B-1 you're right, it's better at modern languages/frameworks and design patterns. Although Opus 4.5 I've found good at everything, even our old C++ projects.

Final thought - Opus 4.5 in claude code CLI was the final moment and mindset change for me. 2026 is going to be wild.
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X