801–810 of 1001 posts
> Spin up agents to tackle routine tasks that take you out of your flow, such as codebase research, bug fixes, and backlog tasks. The software of the future, where nobody on staff knows how anything is built, no one understands why anything breaks, and cruft multiplies exponentially. But at least we're not taken out of our flow!
[flagged]
Earlier quoted context omitted.
Same here - completely relate. One thing I’ve noticed though that actually coding (without the use of AI; maybe a bit of tab auto-complete) is that I’m actually way faster when working in my domain than I am when using AI tools. Everytime I use AI tools in my domain-expertise area, I find it ends up slowing me down. Introducing subtle bugs, me having to provide insane amount of context and details (at which point it…
When I'm doing the coding myself, I'm at least making steady progress and the process is predictable. With LLMs, it's a crapshoot. I have to get the AI to understand what I want and may have try again multiple times, many times never succeeding, and end up writing a lot of text anyway. And in between, I'll have to read a lot of code that probably ends up being thrown away or heavily modified. This probably depends a…
On the pricing page it says that for public preview they are offering a free individual plan with "generous rate limits". I gave it an HTML file and asked it to create Jinja templates from it and 2 minutes later (still planning, no additional prompt) I got this: > Model quota limit exceeded. You have reached the quota limit for this model.
The don't seem to be getting any rate limiting issue which I don't understand, maybe a bug in Antigravity allowing them to use it for more. They are really confident in the IDE after a few hours and the output given is really good.
Earlier quoted context omitted.
I love Zed code. Sad that Sublime Text is not keeping up.
Once the UI soup around AI dev use has settled (and it's getting closer) I bet you we will see native apps with c/c++/zig/rust backends that render so much faster on all the junctions that aren't roundtrip limited (and yes, that will still matter to many people).
I really don't know why I struggle so much with this stuff. I believe these models / agents / whatever write code that is often at least as good as the code I write, and they are super helpful tools, but it just feels like it takes away so much of the joy that is programming to me. I'm not saying it's "right" of me to feel this way, but for me the struggle, and the figuring things out by testing, identifying patterns…
Earlier quoted context omitted.
Why is it so hard for these to be VSCode extensions and not forks?
Microsoft has very specific constraints on what extensions can and can't do, it's not a free for all. They're actively defending their mote by allowing Copilot to do things in a way that extensions couldn't. That's why all the serious contenders make a fork, it's simply not possible to have the same integration otherwise.
I gave it a fair shot. It is a vs code fork. There were some UI glitches. Some usability was better. Cursor has some real annoying usability issues - like their previous/next code change never going away and no way to disable it. Design of this one looks more polished and less muddy. I was working on a project and just continued with it. It was easy because they import setting from cursor. Feels like the browser wars…
The first patch release (released on launch day) says: "Messaging to distinguish particular users hitting their user quota limit from all users hitting the global capacity limits." So, collectively we're hitting the quota, its not just your quota. (One would think Google might know how to scale their services on launch day...) The Documentation ( https://antigravity.google/docs/plans ) claims that "Our modeling sugge…
When I downloaded it, it already came with the proper "Failed due to model provider overload" message.
When it did work, the agent seemed great, achieving the intended changes in a React and python project. Particularly the web app looks much better than what Claude produced.
I did not see functionality to have it test the app in the browser yet.
Earlier quoted context omitted.
These new editors are trying to differentiate themselves via their AI features. Working on the core editor may waste resources that could have been better spent improving the AI features.
Until someone finally figures out that we need to rethink editors from the ground up to support different sort of operations and editing experience, to better facilitate LLMs doing work as agents. But we're probably 1-2 years away from there still, so we'll live with skinned-forks, VSCode extensions and TUIs for now.
Earlier quoted context omitted.
I wouldn't bet on an Electron app winning anything long-term in the dev-oriented space.
I strongly disagree. Firstly, the barrier to entry lower for people to take web experience and create extensions, furthering the ecosystem moat for Electron-based IDEs. Even more importantly, though, the more we move towards "I'm supervising a fleet of 50+ concurrent AI agents developing code on separate branches" the more the notion of the IDE starts to look like something you want to be able to launch in an unconfi…
People dunk on VS Code but it’s pretty damn good. Surely the best Electron app? I’m sure if you are heavily into EMACS it’s great but most people don’t want to invest huge amounts of time into their tools, they would rather be spending that time producing.
For a feature rich workhorse that you can use for developing almost anything straight out of the box, it within minutes after installing a few plugins, it’s very hard to beat. In my opinion lot of the hate is pure cope from people who have probably never really used it.