I've been a software engineer professionally for over two decades and I use AI heavily both for personal projects and at work.
At work the projects are huge (200+ large projects in various languages, C#, TypeScript front-end libs, Python, Redis, AWS, Azure, SQL, all sorts of things).
AI can go into huge codebases perfectly fine and get a root cause + fix in minutes - you just need to know how to use it properly.
Personally I do "recon" before I send it off into the field by creating a markdown document explaining the issue, the files involved, and any "gotchas" it may encounter.
It's exactly the same as I would do with another senior software engineer. They need that information to figure out what is going on.
And with that? They will hand you back a markdown document with a Root Cause Analysis, identify potential fixes, and explain why.
It works amazingly well if you work with it as a peer.
Create a markdown document of your task (or use CLAUDE.md), put it in "plan mode" which allows Claude to use tool calls to ask questions before it generates the plan.
When it finishes one part of the plan, have it create a another markdown document - "progress.md" or whatever with the whole plan and what is completed at that point.
Type /clear (no more context window), tell Claude to read the two documents.
Repeat until even a massive project is complete - with those 2 markdown documents and no context window issues.
> ... Proceeds to explain how it's crippled and all the workarounds you have to do to make it less crippled.
No - that's not what I did.
You don't need an extra-long context full of irrelevant tokens. Claude doesn't need to see the code it implemented 40 steps ago in a working method from Phase 1 if it is on Phase 3 and not using that method. It doesn't need reasoning traces for things it already "thought" through.
This other information is cluttering, not helpful. It is making signal to noise ratio worse.
If Claude needs to know something it did in Phase 1 for Phase 4 it will put a note on it in the living markdown document to simply find it again when it needs it.
Again, you're basically explaining how Claude has a very short limited context and you have to implement multiple workarounds to "prevent cluttering". Aka: try to keep context as small as possible, restart context often, try and feed it only small relevant information.
What I very succinctly called "crippled context" despite claims that Opus 4.5 is somehow "next tier". It's all the same techniques we've been using for over a year now.
I get by because I also have long-term memory, and experience, and I can learn. LLMs have none of that, and every new session is rebuilding the world anew.
And even my short-term memory is significantly larger than the at most 50% of the 200k-token context window that Claude has. It runs out of context before my short-term memory is probably not even 1% full, for the same task (and I'm capable of more context-switching in the meantime).
And so even the "Opus 4.5 really is at a new tier" runs into the very same limitations all models have been running into since the beginning.
> For LLMs long term memory is achieved by tooling. Which you discounted in your previous comments.
My specific complaint, which is an observable fact about "Opus 4.5 is next tier": it has the same crippled context that degrades the quality of the model as soon as it fills 50%.
EMM_386: no-no-no, it's not crippled. All you have to do is keep track across multiple files, clear out context often, feed very specific information not to overflow context.
Me: so... it's crippled, and you need multiple workarounds
scotty79: After all it's the same as your own short-term memory, and <some unspecified tooling (I guess those same files)> provide long-term memory for LLMs.
Me: Your comparison is invalid because I can go have lunch, and come back to the problem at hand and continue where I left off. "Next tier Opus 4.5" will have to be fed the entire world from scratch after a context clear/compact/in a new session.
Unless, of course, you meant to say that "next tier Opus model" only has 15-30 second short term memory, and needs to keep multiple notes around like the guy from Memento. Which... makes it crippled.
If you refuse to use what you call workarounds and I call long term memory then you end up with a guy from Memento and regardless of how smart the model is it can end up making same mistakes. And that's why you can't tell the difference between smarter and dumber one while others can.
I evaluated the claim that Opus is somehow next tier/something different/amazeballs future at its face value. It still has all the same issues and needs all the same workarounds as whatever I was using two months ago (I had a bit of a coding hiatus between beginning of December and now).
> then you end up with a guy from Memento and regardless of how smart the model is
Those models are, and keep being the guy from memento. Your "long memory" is nothing but notes scribbled everywhere that you have to re-assemble every time.
> And that's why you can't tell the difference between smarter and dumber one while others can.
If it was "next tier smarter" it wouldn't need the exact same workarounds as the "dumber" models. You wouldn't compare the context to the 15-30 second short-term memory and need unspecified tools [1] to have "long-term memory". You wouldn't have the model behave in an indistinguishable way from a "dumber" model after half of its context windows has been filled. You wouldn't even think about context windows. And yet here we are
[1] For each person these tools will be a different collection of magic incantations. From scattered .md files to slop like Beads to MCP servers providing access to various external storage solutions to custom shell scripts to ...
BTW, I still find "superpowers" from https://github.com/obra/superpowers to be the single best improvement to Claude (and other providers) even if it's just another in a long serious of magic chants I've evaluated.
> Those models are, and keep being the guy from memento. Your "long memory" is nothing but notes scribbled everywhere that you have to re-assemble every time.
That's exactly how the long term memory works in humans as well. The fact that some of these scribbles are done chemically in the same organ that does the processing doesn't make it much better. Human memories are reassembled at recall (often inaccurately). And humans also scribble when they try to solve a problem that exceeds their short term memory.
> If it was "next tier smarter" it wouldn't need the exact same workarounds as the "dumber" models.
This is akin to opposing calling processor next tier because it still needs RAM and bus to communicate with it and SSD as well. You think it should have everything in cache to be worthy of calling it next tier.
It's fine to have your own standards for applying words. But expect further confusion and miscommunication with other people if don't intend to realign.
> That's exactly how the long term memory works in humans as well.
Where this is applicable when is you go away from a problem for a while. And yet I don't lose the entire context and have to rebuild it from scratch when I go for lunch, for example.
Models have to rebuild the entire world from scratch for every small task.
> This is akin to opposing calling processor next tier because it still needs RAM and bus to communicate with it and SSD as well.
You're so lost in your own metaphor that it makes no sense.
> You think it should have everything in cache to be worthy of calling it next tier.
No. "Next tier" implies something significantly and observably better. I don't. And here you are trying to tell me "if you use all the exact same tools that you have already used before with 'previous tier models' you will see it is somehow next tier".
If your "next tier" needs an equator-length list of caveats and all the same tools, it's not next tier is it?
BTW. I'm literally coding with this "next tier" tool with "long memory just like people". After just doing the "plan/execute/write notes" bullshit incantations I had to correct it:
You're right, I fucked up on all three counts:
1. FileDetails - I should have WIRED IT UP, not deleted it.
It's a useful feature to preview file details before playing.
I treated "unused" as "unwanted" instead of "not yet connected".
2. Worktree not merged - Complete oversight. Did all the work but
didn't finish the job.
3. _spacing - Lazy fix. Should have analyzed why it exists and either
used it or removed the layout constraint entirely.
So next tier. So long memory. So person-like.
Oh. Within about 10 seconds after that it started compacting the "non-crippled" context window and immediately forgot most of what it had just been doing. So I had to clear out the context and teach it the world from the start again.
Edit. And now this amazing next tier model completely ignored that there already exists code to discover network interfaces, and wrote bullshit code calling CLI tools from Rust. So once again it needed to be reminded of this.
> It's fine to have your own standards for applying words. But expect further confusion and miscommunication with other people if don't intend to realign.
I mean, just like crypto bros before them, AI bros do sure love to invent their own terminology and their own realities that have nothing to do with anything real and observable.
> "You're right, I fucked up on all three counts:"
It very well might be that AI tools are not for you, if you are getting such poor results with your methods of approaching them.
If you would like to improve your outcomes at some point, ask people who achieve better results for pointers and try them out. Here's a freebie, never tell AI it fucked up.
"And wow is it nice not to have all the ads and crapware in our faces anymore."
I don't understand this - and I'm not being a Windows defender here, I use Linux when I can (and promote its use).
But my Windows 11 installation has zero ads and zero "crapware". And it's a Dell!
Everything that I didn't want on the machine was removed when I purchased it (two years ago). I see no ads. If I did, this can be fixed easily by even non-technical users with OOShutUp10 or similar - or just edited with a registry change.
I've been using Windows since 3.1 and there were some ugly years but that is not the current state-of-the-state. I'm just calling it like I see it at this point.
The UI is full of Bing and Copilot tie-ins that I consider to be essentially ads. Recommended content in the start menu. The weather widget that shows you news headlines. The lock-screen-of-the-day with the text description that if you accidentally click on it, you open some Bing page. The Edge default home page. Everything is trying to push me towards engaging with Microsoft's online services, which I have never used and have no desire to use. These are ads.
It's probably the case that I could turn all of these off by hunting down the right config options, and if I used Windows as my primary desktop I'm sure I would. But it's just on my game machines which I don't want to spend a lot of time maintaining, and new crap keeps popping up in updates. It's exhausting.
A Debian Linux desktop, in comparison, is not trying to push you to anything. It's a breath of fresh air (not a term I use often but really fits here).
Note: I never made it to Windows 11, only Windows 10. But my understanding is that these things are getting worse, not better. And while not exactly the same thing, there has been a lot of talk lately about how the file explorer has become so bloated and slow that they have to preload it into memory at startup so that it can respond quickly when you click it... omg, I do not want that.
I'm surprised to hear that you were talking about Windows 10. Windows 11 is MUCH worse than 10 with the ads and seems to be the first one where people are complaining en masse. It also comes with a start menu that's both dumbed down and has performance issues. Yes, the start menu. It's slow.
I just know it from a new laptop where I'm keeping the preinstalled Windows for occasions that require it (very rare these days).
The real problem is with trust and encroachment. I think a lot of people that spend a fair amount of time on their computers start to feel like their OS is their home and they go on excursions through apps. Previously, ads were limited to apps you had to go to yourself. Ads showing up as wallpaper in your house would be unsettling, and it reveals that your homeownership was illusory from the start: you never really controlled anything.
Yes, you can use cleanup software to fix the symptoms, but that's not the real issue here.
Edit: further research revealed my original first point was a false assumption.
We must be using different Windows 11 then. Last I booted up Windows instead of shoving Cortana everywhere now it's shoving Copilot. The telemetry sent would make spyware jealous.
The "current" state does not matter. What matters is that MS can shittify your experience at any time. Your machine can stop working if you don't agree to MS "updates". On Linux you have the assurance that the state of your machine can be preserved and you know exactly what's being installed on it.
FTFY: Windows is spyware. The fact that you paid for spyware or it came on your computer or it has useful properties (like Bonzi Buddy) doesn't make it not spyware.
I did a clean Windows 11 install a few months ago. I expected to be bombarded with ads and all of the other things I kept reading about in comments here, but it’s been fine.
I do find it interesting that so many of the comments about how bad Windows 11 is are coming from comments that also admit they aren’t using Windows 11. Not everything in Windows 11 is my favorite design choice, but the anti Windows 11 comments have taken on a life of their own that isn’t always based in reality.
I think people might be equivocating over the word “ad.” Some people consider ads to be interstitial modals which steal focus and have nothing to do with current context. I am much more sensitive and consider any notification to buy or use a service to be an ad. Maybe not pre-installed games but I would prefer they not be there. Microsoft is about as bad as Apple is at suggesting we use their cloud services. I also consider these ads. Still, if I’m honest, they’re infrequent and hardly insurmountable. If one is sensitive to this, the Pro version of Windows makes it easy to disable almost all of this stuff.
they are worst than ads. Onedrive runs automatically and all of a sudden, your files under Documents are all stored by Microsoft without you even knowing.
so you're saying you'd have to download some obscure software just to remove defaults from the OS and that's a plus? The fact that you even know about such software tells me you've already gone thru steps to neuter Microsoft.
> Without the random seed and variable randomness (temperature setting), LLMs will always produce the same output for the same input.
Except they won't.
Even at temperature 0, you will not always get the same output as the same input. And it's not because of random noise from inference providers.
There are papers that explore this subject because for some use-cases - this is extremely important. Everything from floating point precision, hardware timing differences, etc. make this difficult.
You can use Firefox or a Chromium browser that does not have as many of these issues.
I was a Firefox user since the Phoenix/Firebird days but when I wanted Chromium, I chose Brave. It has better blocks for this sort of thing built into it, and uBlock Origin works fine.
It's only the Google Chrome browser that requires the Lite version of that extension. Not Chromium derivatives.
I use Brave + uBlock Origin - problem (for the most part) solved.
There was. New York magazine was the first to report this.
> New York has obtained a confidential document from the Malaysian police investigation into the disappearance of Malaysia Airlines Flight 370 that shows that the plane's captain, Zaharie Ahmad Shah, conducted a simulated flight deep into the remote southern Indian Ocean less than a month before the plane vanished under uncannily similar circumstances.
I've read elsewhere that maybe the cops were finding bits of information (it was just fragment of files, they were not recovered fully) and the whole thing ("He's been planning to fly to the middle of the Indian Ocen!") is just wild speculation backed up with weak "evidence".
Anyway, it's Malaysian authorities. I've lived in the region...
But since we're not citing links, this is all gossip...
It works great. You can put anything you want in there. Coding style, architecture guidelines, project explanation.
Anything the agent needs to know to work properly with your code base. Similar to an onboarding document.
Tools (Claude Code CLI, extensions) will pick them up hierarchically too if you want to get more specific about one subdirectory in your project.
AGENTS.md is similar for other AI agents (OpenAI Codex is one). It doesn't even have to be those - you can just @ the filename at the start of the chat and that information goes in the context.
The naming scheme just allows for it to be automatic.
At work the projects are huge (200+ large projects in various languages, C#, TypeScript front-end libs, Python, Redis, AWS, Azure, SQL, all sorts of things).
AI can go into huge codebases perfectly fine and get a root cause + fix in minutes - you just need to know how to use it properly.
Personally I do "recon" before I send it off into the field by creating a markdown document explaining the issue, the files involved, and any "gotchas" it may encounter.
It's exactly the same as I would do with another senior software engineer. They need that information to figure out what is going on.
And with that? They will hand you back a markdown document with a Root Cause Analysis, identify potential fixes, and explain why.
It works amazingly well if you work with it as a peer.
reply