alexwhb

It’s sadly the new normal. It used to be fast… lately I’ve waited over a week with a small update to my app that has been on the store for over a decade. I hear that it’s likely due to the massive inundation of AI slop apps. But honestly Apple should prioritize apps that have been on for a few years or more.

Ha.. I did the same thing. Massive speed boost for things like this.

Ya migrated about 50% of a code base to a new architecture from a legacy architecture. I've seen a massive speed improvement in my app since doing this and significantly less bugs. The code base is probably on the order of 150,000 lines of code or so. This refactor took about a week with AI... this would have taken easily a month or more if I did this myself. I also had AI write me a bunch of tests that never before existed in this codebase. Obviously these tests aren't perfect, but neither would mine be if I wrote them by hand.

Sure thing. I work on a few projects for my company. The main one is an Android and iOS audiobook-media-player app. I had to update the Android side to use the Google Media3 library instead of the legacy ExoPlayer library. In a typical app this would be pretty straightforward, but we’ve mixed in a lot of custom elements that made the transition quite challenging. I actually attempted it manually back in the day and probably spent three or four days on it, but I simply didn’t have time to finish, so I shelved it. A few months ago I tried again in Codex; within two prompts it was done flawlessly and starting from the beginning.

Another example: I also maintain the back-end API for these apps. There was a lot of old utility code that was messy, unorganized, and frankly gross. I had Codex completely reorganize the code into a sensible folder structure and strip out any functions that weren’t being used anymore—something like 300-plus functions. Doing that by hand would easily have taken a day or more

I could go on, but those were the two initial “aha” moments that showed me how powerful this tool can be when it’s used for the right tasks.

As much as my first gut reaction to this article was to be excited about its conclusion, I can’t say my experience matches up. Are LLMs perfect? Absolutely not, but I can point to many examples in my own work where using Codex has saved me easily a week or more—especially when it comes to tedious refactors. I don’t agree with the conclusion; the real-world improvement and progress I’ve seen in the last year in the problem-solving quality of these agents has been pretty astounding.

The reason for my excitement about the conclusion is obvious: I want programmers and people to stay in demand. I find the AI future to be quite bleak if this tech really does lead to AGI, but maybe that’s just me. I think we’re at a pretty cool spot with this technology right now—where it’s a powerful tool—but at some point, if or when it’s way smarter than you or me… I'm not sure that's a fun happy future. I think work is pretty tied to our self worth as humans.

Ya that’s a solid point. Though many startups give their employees equity options… so you have to factor that in too. Also buying a start up for talent seems risky since many people that join startups are looking for a totally different energy than a large corporation, so it seems reasonable that there’d be a big drop off of that talent as soon as it gets acquired… especially if the vision is not aligned

It’s unclear to me from the article if these volumes had never been discovered before or if the editions of these volumes are unique? Meaning are there wholly unique stories that have been discovered?

Ya honestly that’s a great question. I think more public awareness would be helpful and pressure on state representatives. But honestly you see where it can go badly wrong in Austin TX. A majority (as far as I’m aware) are against the DoT highway expansion and have proposed alternative plans for light rail… but the DoT has over ridden this majority and gone ahead with the highway expansion anyway.

Maybe if the engineers that work on these projects are skilled in planning rail projects too… there’d be less myopic focus on roads. Just a thought.

I think the argument of induced demand is not so much we should never build or expand road infrastructure, that’s obviously asinine, but rather we should consider alternative transportation investments. Rail, dedicated bus lanes etc all increase rider density dramatically over cars. Induced demand will happen regardless of the mode of transport, but cars allow for an extremely low ridership density per lane. That’s really the point being made.

And if you look at the economics of road infrastructure it’s far inferior to other modes when you look at raw cost per person who can and will use it.

Lastly from an environmental, and safety perspective there are several other modes of transport that are far superior

Really interesting read. I’ve had this exact thought but not in a well defined sentence before. Especially regarding entities like turboTax or DoT (Departments of Transportation) where they will expand highways even though it’s a well known empirical fact that this typically causes induced demand or more traffic.

It’s really nice to have such a well defined principle to this idea.

> but oftentimes better than a master of one

I like that. I think that's an accurate statement.

That’s a pretty shocking figure even by tech company standards. I wonder why so much? And how they can justify that?

I find it a bit disheartening that the new wave of “open source” with regards to AI is open weights. That’s like giving someone compiled and obfuscated binaries and saying that’s open source.

I generally agree with this. And so much of modern dev… especially in the app/web dev space is just pattern recognition and hooking up different systems to each other. That’s not to discount its difficulty… but you can be good at several disciplines. I actually think being a generalist (to some degree) can actually be helpful, because it allows you to think in multiple modalities, where as someone who specializes in something is less likely to be able to break out of their thought patterns.