Join the discussion
Write your take first — we'll ask for email only when you're ready to publish.
- Hacker News
- What does any of that mean in practice? it's just rambling about abstract concepts that seem to be designed to hint at a bigger picture, when it's just getting AI to write code for you.
Is this where it's going? Having to mystify our roles so it seems like we're still the thought leaders when actually we're just becoming pseudo-teachers that try and herd our group of AI idiots to the right conclusion for us so we don't have to, without ever giving away that it's just all techno-babble?
by weego - Why is HN interested in management and team process discussion but allergic to similar topics on how to manage agents?
It’s like saying why discuss these team workflows when it’s just devs writing code. Or why use any jargon to describe workflows when it's just devs writing code.
by wahnfrieden - From my past experience of religion at various levels I am very often reminded of borderline-cult religious meetings, and the zeal of converts repeating gnomic oversimplifications, and of how exhausting it was to try to engage with them on any topic of substance.
My own feeling is that it is totally OK to simply route around these people.
It's fascinating how many of the "keep your identity small" folks in the YC/HN sphere have lost any sense of perspective at the first sign of a technology that wanders into the philosophical realm. AI-oriented identities are everywhere.
by dofm - Practically, what this boils down to is having clear success criteria.
The harness (Claude Codex, Codex, Pi etc) keeps throwing things into the context and executing tools (as directed by the model) until the success criteria is satisfied.
The "rules" of using AI successfully are basically just the rules of any successful development team. Break things down into clearly defined chunks, make the success criteria clear and provide a way to get the right feedback on how the system is running (logs, metrics, traces etc).
by bashtoni - > What does any of that mean in practice?
They want you to spend more tokens
by f311a - tech blogs used to read like actionable readme guides. I couldn't finish it without thinking: what am I supposed to do with this information? The shelf life of the latest and greatest is about 2 weeks in the AI space. I never caught up to the ralph wiggum loop and now I'm glad I never tried.by fantasizr
- I'm convinced the field of software engineering is being split in two. These are real concerns and coherent arguments, that make sense for developers who have been using "agentic loops" and heavily AI-assisted workflows.
It scares me that someone can see this as "rambling about abstract concepts" while I see exactly what the author is talking about, both at work and in personal projects; the thought that the majority of people have absolutely no idea what's unfolding.
by ricardobeat - Once you buy into the AI hype, you babble like that. Yegge is an even worse example.by coldtea
- This is not rambling and it is not abstract. The content here is about the second-order effects of "getting AI to write code for you" in less supervised ways. I'll cede the author could have polished this to more concise effect, but at the level it's at, a reader's failure to understand the substance doesn't imply there's mystification going on.by kixiQu
- Thank you for writing this thoughtful post, Armin. I find it deeply comforting that the developer of Pi, an agent harness, does not remove himself from the loop, like me. Maybe if I started thinking of codebases as biological organisms I could get comfortable with getting the human out of the loop.by tmshapland
- I think what is going to happen is revival of "Methodology".
"Methodology" was a big thing in the past just before we got into "Agile Extreme Coding", instead of trying to model the big picture of SW development projects just jump into coding agilly. Implement it feature-by-feature
Granted the methdologies proposed ( See: https://www.ibm.com/docs/en/rational-soft-arch/9.7.0?topic=m... ) may have been too heavy and not flexible and not improved enough. But now with the rise of Agents I think we need to revise and perhaps re-invent them for AI agentic development.
by galaxyLogic - Yes, I believe in reinventing methodologies for the rise of agents.
Before even discussing with my team whether something is a good idea, I can have a full prototype in a new branch within one morning. But then use it as a proof of concept and delete the code.
Another example: agents can generate full test coverage. Including mocking external dependencies and user behavior. Also all in one morning.
Why reinvent methodologies? The cost functions have changed. And agents add new problems, such as hallucinations and tech debt. I think reinventing is a good word since these processes already exist. But the emphasis would shift toward particular parts, making some of them much more mandatory.
by whazor - > My current status is that I have not had much success with this way of working for code I deeply care about
If something is judgement heavy, "code i care deeply about", then i don't really agree with the direction of travel here. Don't try to delegate decisions you care deeply about.
I do like the framing of agent loop vs harness loop, but only delegate stuff that you can accurately specify in advance, that usually means stuff that's repeatable in my case ("hey go see how i did X, do that but for Y"), and that inherently means stuff that's predictable.
For stuff where lack of my judgement as input is just going to cause me to say "no", we're down to collaborating in the "agent loop" as Armin puts it. And that's totally fine. It's fast, but also safe.
Remember before AI coding assistants, sometimes you'd get an engineer join your team who was SUPER productive, your peers would be jealous "oh yeah but you guys only got all that done because you have X on your team!" - they didn't live the curse of having that kind of person around - if you don't have them PERFECTLY aligned, then they run off at break neck speed in the wrong direction.
by CraigJPerry - > Don't try to delegate decisions you care deeply about.
> they didn't live the curse of having that kind of person around - if you don't have them PERFECTLY aligned, then they run off at break neck speed in the wrong direction.
Exactly. If you wouldn't outsource it to people you considered highly skilled, why would you outsource it to a machine?
by zahlman - > Don't try to delegate decisions you care deeply about.
YES. Or find a deterministic way to insert them :D
by otto-riz - I keep thinking about at which point I should not force myself into the loop. As a developer I really like working on the code structure, making it clearer, thinking about good abstraction, breaking into modules, etc. I really take pleasure in it. At the same time I understand that at some point I am becoming the limiting factor.
If the point of the software is benefit people, should I still care about how the code looks.
Right now, I still think that the answer is yes, but in 3 years? in 10 years?
by yanis_t - You will always be able to ask the agent to do refactors for you - and it can do mega ones that exhaust you to think about!by cadamsdotcom
- It's tough if you're somewhere that isn't very meaningful to you beyond the technology. I think there will be an existential shift soon towards more fulfilling work. Maybe I'm naive or that's just what I feel I need for myself.by steezeburger
- > If the point of the software is benefit people, should I still care about how the code looks.
The answer is yes you should, as long as you want to keep software benefiting people.
by wartywhoa23 - > the right fix is not "handle every malformed case." ... [LLMs] will still attempt to handle now impossible errors.
This is the number one code smell from LLMs and I don't know why they are so obsessed with it. In python, it often comes as `hasattr` checks on types that are defined to have that attribute, in a code base that is fully type-checked.
Why do they do that? Is it from pre-training or re-enforcement? If that latter, can the labs please fix this?
by boscillator - Sorry to say but the solution is to stop using python. The models are trained to code defensively assuming historically representative python codebases. The models trust the types a lot more in languages where the canonical historical examples trust the types because the language is constructed around that premise.by CuriouslyC
- million times this - getattr on every dataclass is a wild choiceby efromvt
- It’s because it matches the patterns they are trained to follow. They don’t understand the code. They can’t reason about the actual logic flow. They can only work with patterns.by skywhopper
- Because the vast majority of the codebases in its training set aren't fully type-checked, or very clean at all. Or it's just snippets from Stack Overflow, so there's no existing context to not assume null-checking is valid.by ambicapter
- I suspect it's mostly the training data. I am also on team "make illegal states unrepresentable". It may get talked about a lot on HN, but I'm still at the point that I'm surprised when I see a code base that I didn't write in the wild that does a really good job of it, either open source or at work. Most programmers still think in terms of picking up pieces and fixing errors at the point where the error message pops out rather than making it so the error can't happen and the data reflects that.
I say "mostly" because I think there's also a problem with AIs thinking this way in their current state. That last level of human understanding of a code base, where the human holistically understands the flow of those guarantees, is a challenge to give them right now. On the raw code level, this sort of thing often involves enough code to easily blow out their context window. Trying to summarize it in memories-style files has its own problems; just because there is text written down about the guarantees doesn't mean that the AI is going to get the right info out of it, any more than a human might from just reading the code. I won't say it's "impossible" to give an AI this understanding because I'm not sure it is, but it is a level of understanding of the code that even if you get them to have it, their practices tend to fight against it.
My own solution to this problem has largely been to give up on them getting this. I prompt a solution to the problem the way that most people do, then if I want to make bad illegal states unrepresentable I prompt the AI through the process of the necessary refactorings, unless it's so small that I just do it myself. Given a lot of code that uses maps/dicts and arrays and strings and ints, if you prompt it through making those more thoroughly typed, it's actually pretty good at it. I've not had a lot of luck getting good designs out of single prompts, even when I get detailed. Treating it as two separate tasks seems to work out well.
And watch the diffs on the types carefully; AI loves to sneak past a ".JustSetItAndIgnoreAllThePreAndPostConditions(string)" method. After all, I suspect there's plenty of training data of "types that are nicely structured to make error states unrepresentable and then a later maintainer came along and added a 'JustEffingDoIt' method that broke everything" in the field. One of the best defenses is to make sure that the type implementing these things is in its own file and you can easily look at all the methods it adds on those types and smack it when it does that. I've tried slathering warnings about not doing this and explaining the pre- and post-conditions being maintained in the docs but the change seems marginal.
by jerf - Likely just that they err on the unnecessary error handling than missing error handling. They likely penalize runtime errors harshly in the trainingby rzmmm
- Code is part of a shared and built understanding of an information system.
If these loopers mean we all have to move at this continuous wave of software happening, then we get to the highest levels of logical information system design and its all human judgement and balancing of business requirements to fit a given niche in a company or market. So all the programmers have to become business analysts/market researchers/businessmen...except the specific niches where AI tooling can't really clank well...or the end of the subsidized AI token era makes all this looping too expensive to continue. This feels like expert systems and symbolics lisps machines redux, where we briefly ran into the fact that its not so much the code itself not being able to do stuff, it's that your company's org always gets shipped, so if you can't change your company org, your software only has so much flexibility.
Dataflow diagrams and domain knowledge / domain modeling / ubiquitous languages may become the metalanguage that we start to use and set the standards for quality, functional, and non-functional standards and conventions. We make the "looper clankers" ensure that they fulfill that data / behavior / performance contracts before saying what "done" is, because "done" is no longer just code that compiles, code that builds, code that deploys, or even code that sits in production; it's code that fulfills all of the user requirements, operator requirements, and maintainer requirements. So, the language used may be required to make us all turn into business analysts and software architects more than syntax knowers. The revenge of UML and the return of declarative / logical design / BDD triumphing?
(Typo scan by gemma4-12b but I didn't let it alter my message)
by Multicomp - My experience is that I am bottle-necked on specs. The agent loop is less of a thing for me now.
If I can get a clear understanding of what I want to build, communicate that to Claude Code in planning mode with the goal to write an actionable spec (not code, plan to write the spec) then I tend to get very good results once the agent goes to implement.
But this strategy, while effective, puts a big load on me to write the specs. The agent tends to knock each one out of the park (usually 2 to 3 follow ups based on code review) but then I'm back at the stage that requires the spec.
Another issue for me is that when I step away, if the agent finishes a task and could technically start on an existing spec (no overlap on files so no conflict possible) it doesn't know it can just create a new branch and start. Before I go to bed I'll often say "do task X and once done and pushed start on task Y". But I haven't had luck beyond that. Often I find that it starts on Y and has a question and then the agent is idle the rest of the time.
The final issue is dependency coupled with the above. For example, today I was writing a background job processor. Obviously, the jobs that are in subsequent tasks require the system. That happens with some frequency. Even the specs need to be refreshed after the implementation to take any details that were resolved at coding time into account.
But I am just on the cusp of wanting the outer loop. The gate is almost entirely on spec creation and PR review. In places where those gates don't matter, I want the agent to keep chugging away.
As an aside, I strongly believe we need to start using tools that are better for LLMs even if they are worse for us. For example, Rust is annoying because the compiler is so strict. Bad for me, great for LLMs.
- Why can't Claude write the spec?by andai
- > "Before I go to bed I'll often say "do task X and once done and pushed start on task Y".
Following could solve your problem.
https://www.aihero.dev/skills-handoff
/handoff is my new favourite skill
by Npovview