Astra for Coding: Why Are We Doing This Again?

mrhottakes

> It is absolutely impressive; these types of models are going to change the world in one form or another. > But at least for the moment I don’t know how to work with it for actual software engineering. These two statements seem contradictory.

paxys

It’s time to stop complaining about “shitty code” written by AI. If there are problems in functionality sure, but I personally don’t care about a good directory structure or short, single-purpose functions or short files or whatever else we have been taught about good engineering practices. Why? Because if the code was generated by AI then it will be maintained by AI. I’m not reading it, and I don’t care about it.

data-ottawa

> In fact, with Astra and Fable I feel like not only are the costs astronomical, but the models are also just not for me as a software engineer. This is how I’m feeling as well. The models are so happy to run and implement that they forget to keep me in the loop or make it understandable. If I want to magic something into existence or one shot an analysis, the model is happy. But if I want to direct, understand, and manage they fall completely flat and are unable to explain things for me. It’s even hard to discuss planning because Claude will run off and start pulling data and analyzing my projects and spend 5 minutes on something not needed, when I need a rubber duck that can sanity check things and push back. My non technical colleagues love it, but I’m fighting the thing all the time.

iforgotmypasswo

I have frequently been called a 10x developer by peers. I have written assembly ISRs for wireless firmware. I do full stack development. I’ve reverse engineered proprietary network protocols from dead products for use in specialized communication software. I’ve started a company that’s still running years after I got bored and left. I’ve been hired to start new software departments at companies. I’ve soldered complex electronics together and sold them to companies. At this point I’ve forgotten more interesting projects than I remember. I am very professionally competent. Probably a 5/10 on HN though. Walter Bright and Alan Kay post here sometimes, and you can’t compete with that. I’m not trying to brag. I’m trying to establish that I’m not new to development. AI needs a different architecture and completely changes how to write code that scales well. AI changes what good code looks like because it changes the cost of writing code of a specific quality level altogether. If you are having trouble scaling software development with AI, you need to rethink your software architecture. The more locked in you are to an existing architecture that is not aligned with AI driven development and the less autonomy you have to experiment with new architectures, the more trouble you are going to have. I have PMs doing reliable and high quality frontend development. We now have a substantial number of new tests at the architecture level. All new backend work is in go. I’ve always disliked go as I felt it looks down on ICs, but it’s a great language when you genuinely don’t trust the developers and want to enforce best practices. It wasn’t perfect, but we genuinely made AI driven development work with Opus’s level of intelligence. Some issues, but they were everyday annoyances, not a failure to operate at 5 to 10x team productivity levels. Astra just took that working process and made it 2 to 3x more productive, solving every challenge we were having. It’s nuts.

alex-moon

Benchmaxxing is the word you're reaching for here (though code golf is a perfect name for the symptom). At work I use Codex CLI with GPT 5.4 - it hits a sweet spot: capable enough that it'll do the right thing most of the time, but unsophisticated enough that I can follow along with what it's doing and cut it off and go, "Nope! That's wrong because x, try y instead." For users like me, the threshold for PMF was never "Do I trust it to run unsupervised" but rather "Is supervising it faster than doing it myself" - and as the models get more sophisticated, they actually go backward on this. Most importantly: a bunch of open weights models hit that sweet spot around the same time - I've been having a go with OpenCode and GLM 5.3 on side projects at home and guess what.