171–180 of 1001 posts
Opus 4.5 is incredible, it is the GPT-4 moment for coding because how honest and noticeable the capacity increase is. But still, it has blind spots just like human.
Earlier quoted context omitted.
My company pays for the team Claude code plan which is like $200 a month for each dev. The workflows cost like 10 - 50 cents a PR
It will have to quintuple or more to make business sense for Anthropic. Sure, still cheaper than a full time developer, but don't expect it to stay at $200 for a long time. And then, when you explain to your boss how amazing it is, and can do all this work so easily and quickly, it's when your boss start asking the real question: what am I paying you for?
Why on earth would you be hunting $20 a month subscriptions from random assed people? Peanuts.
Lockheed-Martin could be, but isn’t, opening lemonade stands outside their offices… they don’t because of how buying a Ferrari works.
Opus 4.5 ate through my Copilot quota last month, and it's already halfway through it for this month. I've used it a lot, for really complex code. And my conclusion is: it's still not as smart as a good human programmer. It frequently got stuck, went down wrong paths, ignored what I told it to do to do something wrong, or even repeat a previous mistake I had to correct. Yet in other ways, it's unbelievably good. I ca…
Opus 4.5 really is something else. I've been having a ton of fun throwing absurdly difficult problems at it recently and it keeps on surprising me. A JavaScript interpreter written in Python? How about a WebAssembly runtime in Python? How about porting BurntSushi's absurdly great Rust optimized string search routines to C and making them faster ? And these are mostly just casual experiments, often run from my phone!
> How about porting BurntSushi's absurdly great Rust optimized string search routines to C and making them faster? How did it do? :-)
It couldn't quite beat the Rust implementation on everything, but it managed to edge it out on at least some of the benchmarks it wrote for itself.
(Honestly it feels like a bit of an afront to the natural order of things.)
That said... I'm most definitely not a Rust or C programmer. For all I know it cheated at the benchmarks and I didn't spot it!
Honestly, I don’t understand universal praise for Opus 4.5. It’s good, but really not better than other agents. Just today: Opus 4.5 Extended Thinking designed psql schema for “stream updates after snapshot” with bugs. Grok Heavy gave correct solution without explanations. ChatGPT 5.2 Pro gave correct solution and also explained why simpler way wouldn’t work.
Are you using Claude Code? Because that might be the secret cause you're missing. With Claude Code I can instruct it to validate things after its done with code, and usually it finds that it goofed. I can also tell it to work on like five different things, and go "hey spin up some agents to work on this" and it will spawn 5 agents in parallel to work on said things. I've basically ditched Groke et al and I refuse to…
Logical bug of using BIGSERIAL for tracking updates (generated at insert time, not commit time, so can be out of order) wouldn’t be caught by any number of iterations of Claude Code and would be found in production after weeks of debugging.
Opus 4.5 really is something else. I've been having a ton of fun throwing absurdly difficult problems at it recently and it keeps on surprising me. A JavaScript interpreter written in Python? How about a WebAssembly runtime in Python? How about porting BurntSushi's absurdly great Rust optimized string search routines to C and making them faster ? And these are mostly just casual experiments, often run from my phone!
I'm assuming this refers to the python port of Bellard's MQJS [1]? It's impressive and very useful, but leaving out the "based on mqjs" part is misleading.
Title is: "Opus 4.5 is going to change everything"
Earlier quoted context omitted.
my issue hasn't been for a long time now that the code they write works or doesn't work. My issues all stem from that it works, but does the wrong thing
> My issues all stem from that it works, but does the wrong thing It's an opportunity, not a problem. Because it means there's a gap in your specifications and then your tests. I use Aider not Claude but I run it with Anthropic models. And what I found is that comprehensively writing up the documentation for a feature spec style before starting eliminates a huge amount of what you're referring to. It serves a triple…
This project would have taken me years of specialization and research to do right. Opus's strength has been the ability to both speak broadly and also drill down into low-level implementations.
I can express an intent, and have some discussion back and forth around various possible designs and implementations to achieve my goals, and then I can be preparing for other tasks while Opus works in the background. I ask Opus to loop me in any time there are decisions to be made, and I ask it to clearly explain things to me.
Contrary to losing skills, I feel that I have rapidly gained a lot of knowledge about low-level systems programming. It feels like pair programming with an agentic model has finally become viable.
I will be clear though, it takes the steady hand of an experience and attentive senior developer + product designer to understand how to maintain constraints on the system that allow the codebase to grow in a way that is maintainable on the long-term. This is especially important, because the larger the codebase is, the harder it becomes for agentic models to reason holistically about large-scale changes or how new features should properly integrate into the system.
If left to its own devices, Opus 4.5 will delete things, change specification, shirk responsibilities in lieu of hacky band-aids, etc. You need to know the stack well so that you can assist with debugging and reasoning about code quality and organization. It is not a panacea. But it's ground-breaking. This is going to be my most productive year in my life.
On the flip side though, things are going to change extremely fast once large-scale, profitable infrastructure becomes easily replicable, and spinning up a targeted phishing campaign takes five seconds and a walk around the park. And our workforce will probably start shrinking permanently over the next few years if progress does not hit a wall.
Among other things, I do predict we will see a resurgence of smol web communities now that independent web development is becoming much more accessible again, closer to how it when I first got into it back in the early 2000's.
I see these posts left and right but no one mentions the _actual_ thing developers are hired for, responsibility. You could use whatever tools to aid coding already, even copy paste from StackOverflow or take whole boilerplate projects from Github already. No AI will take responsibility for code or fix a burning issue that arises because of it. The amount of "responsibility takers" also increases linearly with the si…
I'm fine with contributed AI-generated code if someone who's skills I respect is willing to stake their reputation on that code being good.