I agree. The author seems only happy with RAG, insert nail/hammer quote.
I think of contradictory examples where the article doesn't make sense: LLM Chat products are just the product of training data. Generative coding applications take 1 prompt and generate a lot of code.
What this article did make me think is the existing Chat UI in coding apps are too limiting. Some have image attachments, but we need to allow users to input more detail in their prompt (visually, or by having a pre-generation discussion about specifics). That's why I think product engineers will benefit from AI more than non technical folk.
Also, do you have resources that align with those opinions?
Drone startup | Senior Software Engineer, Fullstack Software Engineer | 2 days/week in the office (London, UK) | UK citizens only
You'd be working with other passionate product engineers who previously worked at Google, Amazon, Helsing, Darktrace and Improbable. You'd be working alongside me in building a modern, ground control station application for drones using web technology. For my project, we use React, Tailwind, drizzle-orm, SQLite, Postgres, maplibre, deck.gl, Node, Python, asyncio, FastAPI and more.
Examples of things we're building: low latency data and video streaming services for collaboration, or intuitive/modern applications for planning and performing drone missions.
If interested, please reach out to my temporary disposal email: demise-collar-lurk@duck.com
We highly value passionate engineers with personal projects or technical hobbies. We also value collaboration: pair programming, discussing, teaching and learning from others.
Stealth startup in drones | Senior software engineer, Fullstack software engineer | currently remote, but potentially 2 days/week in the office next year (London, UK) | UK citizens only
You'd be working with other passionate product engineers who previously worked at Google, Amazon, Helsing, Darktrace and Improbable. You'd be working alongside me in building a modern, ground control station software for drones using web technology. For my project, we use React, Tailwind, drizzle-orm, SQLite, Postgres, maplibre, deck.gl, Node, Python, asyncio, FastAPI and more. We highly value passionate engineers with personal projects or technical hobbies, over leetcode engineers. Examples of things we're building: low latency data and video streaming services for collaboration, or intuitive/modern applications for planning and performing drone flights.
If interested, please reach out to my temporary disposal email: whole-heap-justly@duck.com
PS: I value collaboration: pair programming, discussing, teaching and learning from others. One positive signal is that I've gained 27k points on StackOverflow. Bonus points if you've demonstrated positive impact/glue in teams and organisations.
Absolutely! Imaging setting a bunch of css styles through a long winded AI conversation, when you could have an IDE to do it in a few seconds. I don't need that.
The long tail of niche engineering problems is the time consuming bit now. That's not being solved at all, IMHO.
"On the checkout page at the very bottom there are two buttons that are visible when the user chooses to select fast shipping. The right one of those buttons should be a tiny bit more round and it seems like it's not 100% vertically aligned with the other button."
Takes a lot longer to write than just diving into the code. I think that's what they meant.
This checks out with someone I know who grows their own plants, and me, who doesn't spend time growing plants. We're both not interested in the product.
Maybe it has a niche. Millionaires who want to go on holiday but still like to grow plants. It seems more like a gimmick.
I still use StackOverflow. Not as much as I used too, thanks to GPT, but still multiple times a day. What I find is that I spend less time on SO.
However, IMHO deleting questions you originally wrote in the past is hurting other users more than it is hurting AI training.
Other users cannot write similar answers to yours, because it doesn't add anything and they'd get downvoted or deleted. So if you hadn't written your answer years ago, others could've written something similar. Also, other users may have commented on your questions/answers. Their efforts would be lost/deleted if you deleted your questions/answers.
Thanks for your previous contribution to the community. But I would say the worst you should be able to do is remove your name/anonymise your posts, not just delete them.
I wonder would actually deleting questions be a good thing. If there is no old question the same question asked again can not possibly be a duplicate... So constant loop of deleting questions might actually be effective way to fix some problems. And there is enough off-site backups already.
Looking through my browser history, I'd say that I average about 5 distinct SO posts per day. If you know there will be an answer it's less typing to search for it than it is to have ChatGPT regenerate it.
I have tried both. Elysia is nice but one thing I noticed is it absolutely trashed my typescript language server in webstorm, not even using complex routers. Elysia kind of abuses templates, so even simple configurations take a while to process.
I think of contradictory examples where the article doesn't make sense: LLM Chat products are just the product of training data. Generative coding applications take 1 prompt and generate a lot of code.
What this article did make me think is the existing Chat UI in coding apps are too limiting. Some have image attachments, but we need to allow users to input more detail in their prompt (visually, or by having a pre-generation discussion about specifics). That's why I think product engineers will benefit from AI more than non technical folk.
Also, do you have resources that align with those opinions?