I read such articles more or less every day.
This article would be 100% correct if it came out 1 year ago, 75% correct 9 months ago, 50% correct 3 months ago and it's probably 25% correct now if not less.
I totally understand where this is coming from. I too am struggling with accepting that my 30+ years of programming experience is quickly becoming obsolete. I'm losing sleep about this, it's tough.
But just go ahead and give the latest models (Opus 5.5 / Astra 6 as of today) another try.
See what they are capable of and read the code which they produce.
Any problem area, low level C++ or high level Typescript or Clojure or a weird combination of these..
Don't be shy, give them a big task, let them build an entire app, UI and all..
Now compare the output to Opus 4 or gpt-5 from 1 year ago - when they couldn't put together a single function without it being weird and buggy.
This is exactly my problem, not that the models are very good already, but how fast they got so good.
So if coding is not solved yet, it'll get there very soon.
>This article would be 100% correct if it came out 1 year ago, 75% correct 9 months ago, 50% correct 3 months ago and it's probably 25% correct now if not less.
Feels like I read comment similar to this one each year since 2023.
A month ago I was told January of this year was the inflection point. Month before that the inflection point was December of last year. I'm not saying the tech isn't getting better but are the fundamental limitations being surpassed or are the long tail failures just being pushed further away? Because if its the latter this game of "well models really got good nine months ago" won't stop.
It looks to me that this generation of models has reached a level of competence where work can be assigned to them and completed satisfactory for various degrees of satisfactory.
This reflects my own experience with these models. It's not a matter of inflection point if you ask me, it's a matter of accruing capabilities last year the output was not up to my standards 98% of the time, now it looks more like 30% of the time.
I am sure next year models will be better, but the point where the models begun being good enough to start using seriously for my use cases has now passed.
Honestly I feel like 90% of the improvement has been connecting them to deterministic tooling that can tell them when they're wrong. Whenever I chat with one of these latest frontier bots, they still exhibit all the same problems they used to, But throw them in a tight loop with tools that connect to reality, they adjust their behavior accordingly. It's like SLAM vs. dead reckoning for anyone who knows robotics.
temp00345 · · focus · HN ↗
I totally understand where this is coming from. I too am struggling with accepting that my 30+ years of programming experience is quickly becoming obsolete. I'm losing sleep about this, it's tough.
But just go ahead and give the latest models (Opus 5.5 / Astra 6 as of today) another try. See what they are capable of and read the code which they produce. Any problem area, low level C++ or high level Typescript or Clojure or a weird combination of these..
Don't be shy, give them a big task, let them build an entire app, UI and all..
Now compare the output to Opus 4 or gpt-5 from 1 year ago - when they couldn't put together a single function without it being weird and buggy.
This is exactly my problem, not that the models are very good already, but how fast they got so good. So if coding is not solved yet, it'll get there very soon.
lnkl · · focus · HN ↗
Feels like I read comment similar to this one each year since 2023.
throwawayffffas · · focus · HN ↗
That's when the models started to be coherent enough for real work.
They still fuck up, but it does not feel the code was written by drunk interns anymore.
CoolestBeans · · focus · HN ↗
throwawayffffas · · focus · HN ↗
This reflects my own experience with these models. It's not a matter of inflection point if you ask me, it's a matter of accruing capabilities last year the output was not up to my standards 98% of the time, now it looks more like 30% of the time.
I am sure next year models will be better, but the point where the models begun being good enough to start using seriously for my use cases has now passed.
ModernMech · · focus · HN ↗