kaydub

6 ms·

61.
by kaydub 8mo ago
What hard problems are you working on?
62.
by kaydub 8mo ago
I knew nothing about game development a few months ago. Now I've built a simple godot game. I'm sure the game is all pretty common (simple 2d naval combat game) but it's still impressive that a couple claude/gemini/
63.
by kaydub 8mo ago
How long have you been coding? How many languages and frameworks have you worked with? Where has your professional focus been? I've written professional code in production for the past 15+ years in VB, C# (MVC2/MVC3 + Razor), Php(
64.
by kaydub 8mo ago
Always have another agent around that is just reviewing code. "Look for duplicate code, code smells, bad architecture, future scalability problems..." I usually keep one agent/prompt/context ripping through new features
65.
by kaydub 8mo ago
So waterfall sdlc...
66.
by kaydub 8mo ago
Yeah, I'm huge on using LLMs for coding, but one of the biggest wins for me is that the LLM already knows the frameworks. I no longer need to learn whatever newest framework there is. I'll stick to my frameworks, especially when u
67.
by kaydub 8mo ago
I think a lot of the people pro-llm are already in the middle. It's the naysays that are sticking to a strict definition of "vibe coding" as if it's random people typing in "make me $app but make no bugs" and g
68.
by kaydub 8mo ago
> I wonder if the LLMs "converge" on a particular UI if not given specific instructions? Purple. They really fucking like this purple gradient background for some reason lol.
69.
by kaydub 8mo ago
There's a lot of reasons. There's a lot of breadth to "software engineering" (FAANG, web dev, embedded, OS, small business, etc.) I'm sure there are some places where LLMs are bad due to lack of training data. There
70.
by kaydub 8mo ago
Yeah, I don't know why you'd drop using frameworks and libraries just because you're using an LLM. If you AREN'T using them you're just loading a bunch of solved problems into the LLMs context so it can re-invent th
71.
by kaydub 8mo ago
Feels like code written by a co-worker. No different than working on any decent sized code-base anywhere. I've stopped writing code too. Who the fuck wants to learn yet ANOTHER new framework. So much happier with llm tools.
72.
by kaydub 8mo ago
The excitement hasn't cooled off where I'm working. Honestly, I'm personally happy to see so many naysayers online, it means I'm going to have job security a little longer than you folks.
73.
by kaydub 8mo ago
The AI agents can ALREADY "de-slopify" the code. That's one of the patterns people should be using when coding with LLMs. Keep an agent that only checks for code smells, testability, "slop", scalability problems, et
74.
by kaydub 8mo ago
We already do. It's the same tool.
75.
by kaydub 8mo ago
I hope you realize that means your position is in danger.
76.
by kaydub 8mo ago
The hubris is with the devs that think like you actually.
77.
by kaydub 9mo ago
> I have no idea how people use mac. Meh, it has a terminal. Good enough for me. It's worth putting up with MacOS for the hardware.
78.
by kaydub 9mo ago
If there was an easy and supported way to put linux on a macbook I'd be back on linux but I can't give up the hardware.
79.
by kaydub 9mo ago
I haven't even touched a Windows PC in years at this point. Don't know why you'd do that to yourself. I was hanging on to a gaming PC, but it broke during a move over 2 years ago now. I decided that was the last windows box I
80.
by kaydub 9mo ago
I haven't used it because there's been absolutely no reason to even try again. I jumped on copilot when it first came out. I had a personal account and they got us subscriptions at work. It was the first LLM/AI tool I dropped
81.
by kaydub 9mo ago
Rtkit... what are you even working on?
82.
by kaydub 9mo ago
Nope, haven't tried recently because the cli tools were so much better and more powerful I haven't felt like there's a reason to ever go back.
83.
by kaydub 9mo ago
And most the meetings end up with the same people in them where we're just repeating the same shit to an additional 1-2 people per each meeting
84.
by kaydub 9mo ago
I just feel like using Copilot would be like early car designers trying to steer their new car with reins.
85.
by kaydub 9mo ago
I just round robin them until I run out on whatever subscription level I'm on. I only use claude api, so I pay per token there... I consider using claude as "bringing out the big guns" because I also think it's the top-l
86.
by kaydub 9mo ago
Yeah, I've been working with LLMs since openai released that first model. What I'm doing today is VASTLY different than anything we thought possible back then, so I wouldn't call it "vibe coding"
87.
by kaydub 9mo ago
Nope, you couldn't have written it yourself in the same time. That's just a false assumption a lot of you like to make.
88.
by kaydub 9mo ago
I've had success across quite a few languages, more than just python and js. I find it insanely hard to believe you can write code faster than the LLM, even if the LLM has to iterate a couple times. But I'm thankful for you devs t
89.
by kaydub 9mo ago
I don't see people treating AI practices as trade secrets. It's just the nature of a non-deterministic system.
90.
by kaydub 9mo ago
You can update the spec as you go... There's nothing that makes the spec concrete and unchangeable.

More ›