111–120 of 333 posts
Am I the only one disappointed by Github copilot quality lately? I get much better responses by copying and pasting on ChatGPT, rather than the suggestions in the vscode plugin.
I use Copilot in VS code for work and it is great for auto complete, but the chat experience is very poor compared to tools like Cursor. The inline edits are even worse. Long story short, is I don’t have a lot of confidence in the product right now. That being said, I am very optimistic on the product long term and I generally like the vision.
Student here: I legitimately cannot understand how senior developers can dismiss these LLM tools when they've gone from barely stringing together a TODO app to structuring and executing large-scale changes in entire repositories in 3 years. I'm not a singulatarian, but this seems like a brutal S-curve we're heading into. I also have a hard time believing that there is enough software need to make such an extreme prod…
Said it before will say it again, it's a multiplier, that's it.
was able to get around waitlist: copilot-workspace.githubnext.com / / ?task=
Student here: I legitimately cannot understand how senior developers can dismiss these LLM tools when they've gone from barely stringing together a TODO app to structuring and executing large-scale changes in entire repositories in 3 years. I'm not a singulatarian, but this seems like a brutal S-curve we're heading into. I also have a hard time believing that there is enough software need to make such an extreme prod…
Also, the skills you learn as an engineer are highly transferable, as you learn problem solving skills and executive function - many top CEOs have engineering backgrounds. So if you do need to pivot later in your career, you’ll be set up for success
Curious to see how it goes. For 20 years I have been working with Vim and recently added the Co-pilot plugin which is awesome. I can't however see myself using a different editor/IDE than the good'ol vim. It's not that I haven't tried, but all those IDEs tend to be bloated and consume tons of memory.
When GUIs first came out, editors were just "terminal editors in a window". Took a while for the modern concept of an IDE to happen, with hovers, red squigglies, sidebars, jump to definition. All of that was possible on the first day of the GUI editor! But it took a while to figure out what everyone wanted it to be.
I think we're at a similar inflection point. Yeah, everyone today (myself included) is comfortable in the environment we know. VS Code is lovely. And AI (plus realtime multiplayer) is not a display technology. But I think it's a material technology shift in the same vein as those two moments in history. I would not bet that the next thirty years is going to continue look like today's VS Code. I don't know to say what it WILL look like — we have to keep prototyping to find out.
Earlier quoted context omitted.
The hardest part of software development is not writing code, full stop. It never has been and it never will be. The hard part is designing, understanding, verifying, and repairing complex systems. LLMs do not do this, even a little bit.
I guess my worst fear is not "no more jobs because AI can code" but "no more junior jobs because AI can code under the supervision of a senior". SWE jobs will exist, but only seniors will have them and juniors are never hired. Maybe the occasional "apprentice" will be brought on, but in nowhere near the same amount. Where my blind hope lies more specifically is in networking into one of those "apprentice" roles, or m…
Student here: I legitimately cannot understand how senior developers can dismiss these LLM tools when they've gone from barely stringing together a TODO app to structuring and executing large-scale changes in entire repositories in 3 years. I'm not a singulatarian, but this seems like a brutal S-curve we're heading into. I also have a hard time believing that there is enough software need to make such an extreme prod…
I've been asked whether no-code platforms would make us obsolete. I've wondered if quantum computing would make everything we know become obsolete. Now people are wondering whether LLM tools will make us obsolete.
All these things make us more productive. Right now I'm excited by AI tools that are integrated into my IDE and offer to finish my thoughts with a stroke of the 'Tab' key. I'm also very underwhelmed by the AI tools that try to implement the entire project. You seem to be talking about the latter. For the type of coding exercises we do for fun (test drive an implementation of Conway's Game of Life), LLM's are good at them and are going to get better. For the type of coding exercises we do for pay (build a CRUD API), LLM's are mediocre at them. They can give you a starting point, but you're going to do a lot of fiddling to get the schema and business logic right. For the type of coding exercises we do for a lot of pay (build something to solve a new problem in a new way), LLM's are pretty terrible. Without an existing body of work to draw from, they produce code that is either very wrong or subtly flawed in ways that are difficult to detect unless you are an expert in the field.
Right now, they are best used as a productivity enhancer that's autocomplete on steroids. Down the road they'll continue to offer better productivity improvements. But it's very unlikely they will ever (or at least in our lifetimes) entirely replace a smart developer who is an expert in their field. Companies know that the only way to create experts is to maintain a talent pipeline and keep training junior developers in hope that they become experts.
Software development has continued to grow faster than we can find talent. There's currently no indication of LLM's closing that gap.