Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Astra for Coding: Why Are We Doing This Again?

lucumr.pocoo.org

Post reply on HN
Print view View original

281–290 of 334 posts

Re: Astra for Coding: Why Are We Doing This Again?

#281

Post by djeastm »

codingisfreedom wrote: post #43

I’ve asked Astra to build me an app for a prototype I created quickly using Sonnet. It’s been 2 days and it made no real progress on the actual app. It created docs, scripts, workflows, and it’s doing a bunch of reviewing on every PR. I told it that I just need an MVP. I’m pretty sure an average senior engineer would have finished that task much quicker, and guaranteed with more readable, higher-quality code. Meanwhi…

>I think I’ve easily crossed 100k tokens so far on nothing

100k tokens? Is it just me or is that very low for an app build?

Re: Astra for Coding: Why Are We Doing This Again?

#282

Post by djeastm »

AmazingTurtle wrote: post #29

gpt-6-astra is a bitch, it constantly scope creeps itself with "yet another thing" to give it that darn polished lick. the results are eventually a little bit better but at what cost? let's do the math. gpt-5.6-sol: 1x base gpt-6-astra 2.5x base in subscription then gpt-6-astra tends to spawn subagents a lot, often with all kinds of models such as gpt-5.6, 5.3-codex etc., which is neat. it's a good coordinator but ev…

I might be missing something here, but can't you just put in AGENTS.md something like "do not run full test suite unless asked" or something?

Re: Astra for Coding: Why Are We Doing This Again?

#283

Post by officialchicken »

iforgotmypasswo wrote: post #279
taurath wrote: post #120

When the code is shitty it becomes harder and harder for the models to make changes and this grinds progress down to a halt - this has been my experience with “factories” trying them and doing refining steps every few months. I sincerely don’t understand what the people who say they no longer read any code are doing, because it must be somewhat trivial to not run headlong into these issues that stack up time after ti…

I’ll try to explain how to do it correctly. I’m not selling anything. Seeing this as the top comment makes me a bit sad. 1. Learn about ports and adapters as an architecture pattern. Domain driven design and locality of reasoning are your new best friends. 2. Realize that AI can generate unlimited fake data almost immediately. So anything you can isolate can get a fake adapter and a real one. You can build and test a…

I can't agree with 4 - it's sophomoric reasoning at it's best. The code is the product, it's what the system (human/ai/factory/combo/etc) is producing. The IC will always be more familiar with the nuance and the implications of the decisions than the manager. There is only one real stat to track - profit. As for the size of your team, not all human developers are equal, but agentic tend to behave similarly. A small team of highly coordinated things will always outproduce a pile of generic ones acting will little or no methodology. Please deeply re-evaluate at a philosophical level what quality over quantity really means for delivering outcomes.

Re: Astra for Coding: Why Are We Doing This Again?

#284

Post by VulgarExigency »

samuell wrote: post #138
gps372 wrote: post #55

Early lesson I learned from AI engineering was - there is no substitute to giving a groomed epic to an agent. Instead of simply saying 'implement themes in my product' you need to be specific, in fact more specific than usual. You need to say exactly what is in scope and what's not, even down to a buttons, events and layouts. You can groom the epic with the help of AI, but final review must be done by someone who can…

Yeah, I think as a general direction we need to allow a useful collaboration with the agents on the issue board. This is btw why Epiq was developed, to keep the board as code, git-backed, distributed (via an event log mechanism), and with the ability to replay the board, to see what agents actually did: https://ljtn.github.io/epiq

How does this compare to Beads, if you don't mind me asking?

https://github.com/gastownhall/beads

Re: Astra for Coding: Why Are We Doing This Again?

#285

Post by thatjoeoverthr »

taurath wrote: post #120

When the code is shitty it becomes harder and harder for the models to make changes and this grinds progress down to a halt - this has been my experience with “factories” trying them and doing refining steps every few months. I sincerely don’t understand what the people who say they no longer read any code are doing, because it must be somewhat trivial to not run headlong into these issues that stack up time after ti…

While there are ways to manage it with supervision, the impulses of every model are clear and counterproductive. All code is littered CoT in comments and insane decisions. Its attention is limited and it saturates fast, neutralizing one of its main advantages. It’s really nice with supervision but something smells with the hustlers.

Re: Astra for Coding: Why Are We Doing This Again?

#286

Post by zamadatix »

amai wrote: post #275

The idea to use python code instead of other kinds of tool calls is taken from smolagents: https://huggingface.co/blog/smolagents It is based on this paper https://huggingface.co/papers/2402.01030 and calls this idea CodeAct. The paper is actually from Apple: https://machinelearning.apple.com/research/codeact So Astra and Fable seem to take this idea to the extreme causing some unwanted side-effects.

I'm not sure the idea is really from a single set place or lineage like that.

If it was, it was at least from before smolagents and those papers - ChatGPT had already been using automatic Python scripting+evaluation calls and people calling it in agentic loops in 2023. The ReAct paper for agentic loops and PAL paper for dynamically calling Python for tasks which can be better done computationally were both from 2022 (but that doesn't mean the idea necessarily sprung from those either, they're just earlier papers published on the topics).

Re: Astra for Coding: Why Are We Doing This Again?

#287

Post by nchie »

defmacr0 wrote: post #250
nchie wrote: post #173

Earlier quoted context omitted.

For my side project where uptime doesn't really matter but the codebase is quite huge (gamedev of a multiplayer game with Bevy, so both server code, game code, content build pipeline, as well as a TON of editor and authoring code, probably 150k+ LoC), I've been moving towards just reviewing boundaries between modules, APIs, and schemas, and I'm pushing towards keeping things as modular as possible. If something ends…

This can easily lead you down a path of excessive modularity with each module being excessively complex though

In my experience, when that happens it's most likely because you drew the wrong boundaries. Iterating on the boundaries also becomes quite cheap when developing this way though, you should never expect to get them right the first time unless it's a very common problem you're solving _or_ you've done it before.

Re: Astra for Coding: Why Are We Doing This Again?

#288

Post by vincston »

troupo wrote: post #22
idiliv wrote: post #15

Earlier quoted context omitted.

What is the "compilers argument"?

People keep saying that "models are just compilers, and I don't see you complsining about compilers". Which is such a bullshit argument

Can you elaborate? Is it because the models are inherently not deterministic? But then, will it not get better with time? I mean arent we just at the beginning of the research here?

Re: Astra for Coding: Why Are We Doing This Again?

#289

Post by datsci_est_2015 »

LtWorf wrote: post #277
wallst07 wrote: post #237

Earlier quoted context omitted.

Or C, agentic coders who have been doing this for 20+ years (software engineering) know what they're doing? Also, when people say 'read the code' do they really mean go line by line, or review the pseudo code? Meaning, read the high level architecture/data flow. Because IMO the code matters very little if you have the proper testing environment and guardrails, the architecture always matters.

In my circle, the only people who make these claims about AI were all terrible developers whose contributions were slowing down everyone because they needed to go through several reviews and rewrites before they could be accepted. Now it's basically the same but they love it because it's only their teammates who have to put in the work not them. This all implodes when the teammates get fed up and just approve everyth…

+1

Some people write terrible code. Some people don’t proofread their own code. Some people are writing code in a second (or third!) language and the typos are harder to spot and comments are harder to formulate.

It’s a bit like the variance you could expect from asking a room full of people to write a 1-page short story on a specific topic (e.g. “death of a loved one”). Some of those short stories will be unreadable, and some of the people who wrote those terrible stories will have no idea how bad they are. Except now, enter AI, and the room generates the short story instead - who has the skills to determine whether their short story isn’t terrible?

Re: Astra for Coding: Why Are We Doing This Again?

#290

Post by datsci_est_2015 »

whstl wrote: post #195
Gigachad wrote: post #100

Earlier quoted context omitted.

They don’t want to sell these tools to developers. They want to cut as many layers as possible.

Where I work: Developers very rarely blow their limits, except when they're experimenting on purpose. Most non-developers are out of tokens by the half of the week, and need to use usage credits for the remainder. To me there is clearly a better target demographic for AI.

This is a very insightful dynamic.

Probably reinforces that we’ve already surpassed the frontier threshold for LLM usability in software development and can now focus on cost and personalization. To make a comparison, no one is making a better machine vision app for hot dog classification - we hit diminishing returns 10 years ago on that front.

But also scary for both investors and the working class: AI companies want to facilitate the concentration of capital even further into the hands of the ownership class. Will they succeed?

Post reply on HN