Workflow and Productivity Patterns

Developers sharing how they use AI as a pairing partner, strategies for maintaining code structure while using AI, using AI for guidance rather than autonomous decision-making, and treating AI like a junior developer

Many developers treat AI not as an autonomous decision-maker but as an "eager junior developer" that requires strict architectural guidance and "human-in-the-loop" supervision to avoid low-quality code. By stubbing out data models and requiring design documents before any code is generated, experienced users maintain control over long-term maintainability while using the AI as a high-speed typist to handle repetitive boilerplate. This "programmer mindset"—prioritizing structured responsibilities and precise file management—is increasingly being applied to general knowledge work, enabling users to automate complex, cross-application workflows like social media scheduling and email management. Ultimately, visibility into an AI’s reasoning process remains a critical feature for these power users, as it allows them to course-correct and ensure the agent doesn't drift into logical errors or "confused" assumptions.

View on HN · Topics

> There seems a fair enthusiasm in the UI of these to hide code from coders. Like the prompt interaction is the true source and the actual code is some sort of annoying intermediate runtime inconvenience to cover up.

I've finally started getting into AI with a coding harness but I've take the opposite approach. usually I have the structure of my code in my mind already and talk to the prompt like I'm pairing with it. while its generating the code, I'm telling it the structure of the code and individual functions. its sped me up quite a lot while I still operate at the level of the code itself. the final output ends up looking like code I'd write minus syntax errors.

View on HN · Topics

This is the way to do it if you're a serious developer, you use the AI coding agent as a tool, guiding it with your experience. Telling a coding agent "build me an app" is great, but you get garbage. Telling an agent "I've stubbed out the data model and flow in the provided files, fill in the TODOs for me" allows you the control over structure that AI lacks. The code in the functions can usually be tweaked yourself to suit your style. They're also helpful for processing 20 different specs, docs, and RFCs together to help you design certain code flows, but you still have to understand how things work to get something decent.

Note that I program in Go, so there is only really 1 way to do anything, and it's super explicit how to do things, so AI is a true help there. If I were using Python, I might have a different opinion, since there are 27 ways to do anything. The AI is good at Go, but I haven't explored outside of that ecosystem yet with coding assistance.

View on HN · Topics

My workflow is quite similar. I try to write my prompts and supporting documentation in a way that it feels like the LLM is just writing what is in my mind.

When im in implementation sessions i try to not let the llm do any decision making at all, just faster writing. This is way better than manually typing and my crippling RSI has been slowly getting better with the use of voice tools and so on.

View on HN · Topics

Do you ask it for a design first? Depending on complexity I ask for a short design doc or a function signature + approach before any code, and only greenlight once it looks sane.

View on HN · Topics

Because it's not "PhD-expert level" at all, lol. Even the biggest models (Mythos, GPT-Pro, Gemini DeepThink) are nowhere near the level of effort that would be expected in a PhD dissertation, even in their absolute best domains. Telling it to work out a plan first is exactly how you would supervise an eager but not-too-smart junior coder. That's what AI is like, even at its very best.

View on HN · Topics

> the code is not abstracted, it's an important artifact and it needs inspection.

That's a rather trivial consideration though. The real cost of code is not really writing it out to begin with, it's overwhelmingly the long-term maintenance. You should strive to use AI as a tool to make your code as easy as possible to understand and maintain, not to just write mountains of terrible slop-quality code.

View on HN · Topics

I want it yes. I already feel like Im the one doing the dumb work for the AI of manually clicking windows and typing in a command here or there it cant do.

Ive also been getting increasingly annoyed with how tedious it is to do the same repetitive actions for simple tasks.

View on HN · Topics

This is me!

I’m semi-normie (MechEng with a bit of Matlab now working as a ceo).

I spend most of my day in Claude code but outputs are word docs, presentations, excel sheets, research etc.

I recently got it to plan a social media campaign and produce a ppt with key messaging and content calendar for the next year, then draft posts in Figma for the first 5 weeks of the campaign and then used a social media aggregator api to download images and schedule in posts.

In two hours I had a decent social media campaign planned and scheduled, something that would have taken 3-4 weeks if I had done it myself by hand.

I’ve vibe coded an interface to run multiple agents at once that have full access via apis and MCPs.

With a daily cron job it goes through my emails and meeting notes, finds tasks, plans execution, executes and then send me a message with a summary of what it has done.

Most knowledge work output is delivered as code (e.g. xml in word docs) so it shouldn’t be that that surprising that it can do all this!

View on HN · Topics

I am starting to use Codex heavily on non-coding tasks. But I am realizing it works because I work and think like a programmer - everything is a file, every file and directory should have very precise responsibilities, versioning is controlled, etc. I don't know how quick all of this will take to spread to the general population.

View on HN · Topics

Most knowledge workers aren't willing to put in the effort so they're getting their work done efficiently.

View on HN · Topics

> The part that makes this powerful is that the LLM is the ultimate UI/UX.

I strongly doubt that. That’s like saying conversation is the ultimate way to convey information. But almost every human process has been changed to forms and structured reports. But we have decided that simple tools does not sell as well and we are trying to make workflow as complex as possible. LLM are more the ultimate tools to make things inefficient.

View on HN · Topics

Yes, I have a theory - that higher efficiency becomes structural necessity. We just can't revert to earlier inefficient ways. Like mitochondria merging with the primitive cell - now they can't be apart.

View on HN · Topics

it was many hours of working with codex, guidance and comparing to known-good outputs from previous years, but a sufficiently smart model would be able to just do it without any steering; it'd still take hours, but my input wouldn't be necessary. a harness for getting this done probably exists today, gastown perhaps or something that the frontier labs are sitting on.

View on HN · Topics

Just got Computer Use working and honestly it feels really, really good. This is going to enable so many high-quality cross-application workflows in non-browser applications.

View on HN · Topics

Couple of people in my company have vibe coded some chat interface and they’re passing skills and MCPs that give the model access to all our internal data (multiple databases) and tools (Jira, Confluence etc).

I wonder if there’s something off the shelf that does this?

View on HN · Topics

Slides, publications and tech reports, very handy for figures !

View on HN · Topics

I think the killer feature in this release is the background GUI use.

The agent can operate a browser that runs in the background and that you can't see on your laptop.

This would be immensely useful when working with multiple worktrees. You can prompt the agent to comprehensively QA test features after implementing them.

View on HN · Topics

First use case I'm putting to work is testing web apps as a user. Although it seems like this could be a token burner. Saving and mostly replaying might be nice to have.

View on HN · Topics

The macOS app version of Codex I have doesn't show reasoning summaries, just simply 'Thinking'.

Reasoning deltas add additional traffic, especially if running many subagents etc. So on large scale, those deltas maybe are just dropped somewhere.

Saying that, sometimes the GPT reasoning summary is funny to read, in particular when it's working through a large task.

Also, the summaries can reveal real issues with logic in prompts and tool descriptions+configuration, so it allowing debugging.

i.e. "User asked me to do X, system instructions say do Y, tool says Z which is different to what everyone else wants. I am rather confused here! Lets just assume..."

It has previously allowed me to adjust prompts, etc.

View on HN · Topics

It's useful when using prism, and for exploratory research & code.

View on HN · Topics

I do want to see as it allows me to course correct.

View on HN · Topics

at least for our scope of work (data, interfacing with data, building things to extract data quickly and dump to warehouse, resuming) claude is performing night and day better than codex. we're still continuing tinkering with codex here to see if we're happy with it but it's taking a lot more human-in-the-loop to keep it from going down the wrong path and we're finding that we're constantly prompt-nudging it to the end result. for the most part after ~3 days we're not super happy with it. kinda feels like claude did last year idk. it's worth checking out and seeing if it's succeeding at the stuff you want it to do.

Summarizer