What's the largest software project AI can complete on its own?

1. You have very good tests. The agent can write them, but you have to prompt it to do that, and some human input may be needed here.

2. You have a good, modular architecture. Again, this is something that needs to be prompted and/or developed up front. Agent work can scale if you divide and conquer, so that any given task has a well-defined scope, limiting the amount of context needed to handle it well. If you let an agent just build a big pile of code, you’ll have issues.

Once you have the above, you’re more likely to be able to just let the agent run.

This is why you hear about things like SQLite being ported to Rust (by multiple groups!) It has a comprehensive unit test suite and a good existing architecture. Such porting projects are benefitting from already having done the upfront work that I mentioned, plus they have the original source code as a reference implementation.

Because of that, most of what I do involves a lot of interaction with the agent. That’s because I’m typically working on complex existing systems, that don’t necessarily have the above properties, and agents aren’t at the point where they can handle that autonomously.

That isn't true. Everyone knows it and you know it. So why would you make a statement like this? Does it make you feel like a "thought leader"? This isn't X or LinkedIn. Bulshitting here just makes you look like a fool.

I don't think i wrote more than 10 lines of code in the largest project i'm working on. Lines of code: Java: 900_635, typescript: 725_418, C++: 180_445, Dart: 96_181.

It's been obvious from the start that no model, as good as it is, can do large(-ish) amounts of work by its own without supervision, control, criticism, etc ... If left unsupervised, models usually do half the work, leaving stubs and todos everywhere.

Quality comes from applying software engineering principles as much as possible, just like you would do with teams of junior devs: planning sessions and implementation sessions with adversarial critiques, specifying as much as possible upfront, planning unit/smoke/integration tests, etc ...

Many systems rely on swarm of agents to build software but i've found it very difficult to get good results without lots of overhead/token waste because of inter agent communications mostly.

So instead i built what is mostly a workflow engine to structure / organize processes into workflows with different agents assigned different roles. I've setup a basic landing page here https://kodfactory.com if anyone wants to follow along.

However, in the case of the apt cacher, the DDoS of Ubuntu infrastructure showed that the existing apt cachers left a *LOT* of room for improvement, not to mention that apt-cacher-ng has regularly (over the last 13 years of use) gotten wedged and needed intervention.

This cacher understands Deb package repo structure, and when it finds new meta-data it grabs and checks the meta-data, downloads new version of any package you regularly download (configurable), and then makes a snapshot of that, which it then promotes for cache users to see.

I can shut down my Internet connection and then install a fresh Ubuntu from ISO, then "apt update; apt dist-upgrade" successfully via the cache.

If you are saying that AI isn't solving novel problems, you are waaaaaay off.

But you're also missing the point: This branch of the conversation is about AI tools writing wildly inflated code. Which I proposed a few data-points refuting.