Opus 4.5 is not the normal AI agent experience that I have had thus far • Hacker News Board

Re: Opus 4.5 is not the normal AI agent experience that I have had thus far

#321

Post by sothatsit » Wed, Jan 07, 2026, 12:34 AM UTC

honeycrispy wrote: post #26

A couple weeks ago I had Opus 4.5 go over my project and improve anything it could find. It "worked" but the architecture decisions it made were baffling, and had many, many bugs. I had to rewrite half of the code. I'm not an AI hater, I love AI for tests, finding bugs, and small chores. Opus is great for specific, targeted tasks. But don't ask it to do any general architecture, because you'll be soon to regret it.

I like these examples that predictably show the weaknesses of current models.

This reminds me of that example where someone asked an agent to improve a codebase in a loop overnight and they woke up to 100,000 lines of garbage [0]. Similarly you see people doing side-by-side of their implementation and what an AI did, which can also quite effectively show how AI can make quite poor architecture decisions.

This is why I think the “plan modes” and spec driven development are so important effective for agents, because it helps to avoid one of their main weaknesses.

[0] https://gricha.dev/blog/the-highest-quality-codebase

Re: Opus 4.5 is not the normal AI agent experience that I have had thus far

#322

Post by dbbk » Wed, Jan 07, 2026, 12:36 AM UTC

ryanchants wrote: post #281

lelanthran wrote: post #23

Earlier quoted context omitted.

> You have to experience it yourself on your own real problems and over the course of days or weeks. How do you stop it from over-engineering everything?

I have it propose several approaches, pick and choose from each, and remove what I don't want done. "Use the general structure of A, but use the validation structure of D. Using a view translation layer is too much, just rely on FastAPI/SQLModel's implicit view conversion."

The Plan mode already does this, it makes multiple plans and then synthesises them

Re: Opus 4.5 is not the normal AI agent experience that I have had thus far

#323

Post by whatever1 » Wed, Jan 07, 2026, 12:36 AM UTC

multisport wrote: post #320

What bothers me about posts like this is: mid-level engineers are not tasked with atomic, greenfield projects. If all an engineer did all day was build apps from scratch, with no expectation that others may come along and extend, build on top of, or depend on, then sure, Opus 4.5 could replace them. The hard thing about engineering is not "building a thing that works", its building it the right way, in an easily unde…

Their thesis is that code quality does not matter as it is now a cheap commodity. As long as it passes the tests today it's great. If we need to refactor the whole goddamn app tomorrow, no problem, we will just pay up the credits and do it in a few hours.

Re: Opus 4.5 is not the normal AI agent experience that I have had thus far

#324

Post by multisport » Wed, Jan 07, 2026, 12:38 AM UTC

whatever1 wrote: post #323

multisport wrote: post #320

What bothers me about posts like this is: mid-level engineers are not tasked with atomic, greenfield projects. If all an engineer did all day was build apps from scratch, with no expectation that others may come along and extend, build on top of, or depend on, then sure, Opus 4.5 could replace them. The hard thing about engineering is not "building a thing that works", its building it the right way, in an easily unde…

Their thesis is that code quality does not matter as it is now a cheap commodity. As long as it passes the tests today it's great. If we need to refactor the whole goddamn app tomorrow, no problem, we will just pay up the credits and do it in a few hours.

Yes agreed, and tbh even if that thesis is wrong, what does it matter?

Re: Opus 4.5 is not the normal AI agent experience that I have had thus far

#325

Post by klaussilveira » Wed, Jan 07, 2026, 12:38 AM UTC

OldGreenYodaGPT wrote: post #33

Most software engineers are seriously sleeping on how good LLM agents are right now, especially something like Claude Code. Once you’ve got Claude Code set up, you can point it at your codebase, have it learn your conventions, pull in best practices, and refine everything until it’s basically operating like a super-powered teammate. The real unlock is building a solid set of reusable “skills” plus a few agents for th…

I made a similar comment on a different thread, but I think it also fits here: I think the disconnect between engineers is due to their own context. If you work with frontend applications, specially React/React Native/HTML/Mobile, your experience with LLMs is completely different than the experience of someone working with OpenGL, io_uring, libev and other lower level stuff. Sure, Opus 4.5 can one shot Windows utilities and full stack apps, but can't implement a simple shadowing algorithm from a 2003 paper in C++, GLFW, GLAD: https://www.cse.chalmers.se/~uffe/soft_gfxhw2003.pdf

Codex/Claude Code are terrible with C++. It also can't do Rust really well, once you get to the meat of it. Not sure why that is, but they just spit out nonsense that creates more work than it helps me. It also can't one shot anything complete, even though I might feed him the entire paper that explains what the algorithm is supposed to do.

Try to do some OpenGL or Vulkan with it, without using WebGPU or three.js. Try it with real code, that all of us have to deal with every day. SDL, Vulkan RHI, NVRHI. Very frustrating.

Try it with boost, or cmake, or taskflow. It loses itself constantly, hallucinates which version it is working on and ignores you when you provide actual pointers to documentation on the repo.

I've also recently tried to get Opus 4.5 to move the Job system from Doom 3 BFG to the original codebase. Clean clone of dhewm3, pointed Opus to the BFG Job system codebase, and explained how it works. I have also fed it the Fabien Sanglard code review of the job system: https://fabiensanglard.net/doom3_bfg/threading.php

We are not sleeping on it, we are actually waiting for it to get actually useful. Sure, it can generate a full stack admin control panel in JS for my PostgreSQL tables, but is that really "not normal"? That's basic.

Re: Opus 4.5 is not the normal AI agent experience that I have had thus far

#326

Post by christophilus » Wed, Jan 07, 2026, 12:40 AM UTC

minimaxir wrote: post #73

nikisil80 wrote: post #11

[flagged]

> Re-building things that most probably already exist, simply with your own little special flavour? That describes half of the current unicorn startups nowadays.

More than half. What has anyone written that was truly new? Regardless, if you have an idea, you will build it out of some combination of conditionals, loops, and expressions… turns out agents are pretty good at those things, even when the idea you’re expressing is novel.

Re: Opus 4.5 is not the normal AI agent experience that I have had thus far

#327

Post by simonw » Wed, Jan 07, 2026, 12:41 AM UTC

mattfrommars wrote: post #301

I’ve been saying this a countless time, LLM are great to build toy and experimental projects. I’m not shaming but I personally need to know if my sentiment is correct or not or I just don’t know how to use LLMs Can vibe coder gurus create operating system from scratch that competes with Linux and make it generate code that basically isn’t Linux since LLM are trained on said the source code … Also all this on $20 plan…

Your bar for being impressed by coding agents is "can build a novel operating system that competes with Linux on a plan that costs a $20/month"?

Yeah, they can't do that.

Re: Opus 4.5 is not the normal AI agent experience that I have had thus far

#328

Post by jasonfarnon » Wed, Jan 07, 2026, 12:43 AM UTC

TacticalCoder wrote: post #255

OldGreenYodaGPT wrote: post #33

Most software engineers are seriously sleeping on how good LLM agents are right now, especially something like Claude Code. Once you’ve got Claude Code set up, you can point it at your codebase, have it learn your conventions, pull in best practices, and refine everything until it’s basically operating like a super-powered teammate. The real unlock is building a solid set of reusable “skills” plus a few agents for th…

> Most software engineers are seriously sleeping on how good LLM agents are right now, especially something like Claude Code. Nobody is sleeping. I'm using LLMs daily to help me in simple coding tasks. But really where is the hurry? At this point not a few weeks go by without the next best thing since sliced bread to come out. Why would I bother "learning" (and there's really nothing to learn here) some tool/workflow…

"But really where is the hurry?" It just depends on why you're programming. For many of us not learning and using up to date products leads to a disadvantage relative to our competition. I personally would very much rather go back to a world without AI, but we're forced to adapt. I didn't like when pagers/cell phones came out either, but it became clear very quickly not having one put me at a disadvantage at work.

Re: Opus 4.5 is not the normal AI agent experience that I have had thus far

#329

Post by whatever1 » Wed, Jan 07, 2026, 12:45 AM UTC

multisport wrote: post #324

whatever1 wrote: post #323

Earlier quoted context omitted.

Their thesis is that code quality does not matter as it is now a cheap commodity. As long as it passes the tests today it's great. If we need to refactor the whole goddamn app tomorrow, no problem, we will just pay up the credits and do it in a few hours.

Yes agreed, and tbh even if that thesis is wrong, what does it matter?

The whole point of good engineering was not about just hitting the hard specs, but also have extendable, readable, maintainable code.

But if today it’s so cheap to generate new code that meets updated specs, why care about the quality of the code itself?

Maybe the engineering work today is to review specs and tests and let LLMs do whatever behind the scenes to hit the specs. If the specs change, just start from scratch.

Re: Opus 4.5 is not the normal AI agent experience that I have had thus far

#330

Post by wild_egg » Wed, Jan 07, 2026, 12:46 AM UTC

klaussilveira wrote: post #325

OldGreenYodaGPT wrote: post #33

Most software engineers are seriously sleeping on how good LLM agents are right now, especially something like Claude Code. Once you’ve got Claude Code set up, you can point it at your codebase, have it learn your conventions, pull in best practices, and refine everything until it’s basically operating like a super-powered teammate. The real unlock is building a solid set of reusable “skills” plus a few agents for th…

I made a similar comment on a different thread, but I think it also fits here: I think the disconnect between engineers is due to their own context. If you work with frontend applications, specially React/React Native/HTML/Mobile, your experience with LLMs is completely different than the experience of someone working with OpenGL, io_uring, libev and other lower level stuff. Sure, Opus 4.5 can one shot Windows utilit…

I've had Opus 4.5 hand rolling CUDA kernels and writing a custom event loop on io_uring lately and both were done really well. Need to set up the right feedback loops so it can test its work thoroughly but then it flies.