Ask HN: By what percentage has AI changed your output as a software engineer?

Compared to the era before AI coding tools (say, ~2 years ago), if you had to put a number on it, how much has your productivity as a SWE changed?

I think about this a lot; am keen to hear what others' perceptions are. For me; the short answer: about 2x (i.e. 100% faster than pre LLMs). Long answer:

When I thoroughly understand the domain (i.e. business logic and real world problem I'm solving), and am familiar with the tech stack, I'm about ~10x faster for the same or better code quality.

When I don't understand the domain, prompts will be ambiguous or inadequate, the LLM will guess, it will do a month's work in a day, but I'll spend the next 3 weeks refactoring and realising how trash the code was, due to how trash the prompt was. All in all, it's probably still faster than pre AI, but can give a demoralising psychological phenomena where you think something's nearly completed only to spend weeks debugging it, refactoring, and often tossing it away and starting over.

In an unfamiliar tech stack, I can't always spot obvious mistakes (mistakes caused by the AI or the prompt), so it's less productive and more risky.

10-15% of the productivity improvement is due to improvements in the dev environment. I open ~/dotfiles with cursor and tell it a problem I have or ask for a specific improvement. It usually modifies .zshrc, .vimrc or similar (and iterates as necessary if the first attempt didn't work). Due to how fast this is (e.g. 5 minutes), I've made about 20 little tweaks that previously didn't justify the time. They definitely make me happier as well as a bit more productive.

But overall, taking everything into account, I'd say I'm about 2x as productive as before LLMs.

I have yet to see a study design that sufficiently controls for all of the variables. In general it seems that if you could do the work on your own, it may not save time, and in the long run, most all output needs to be reviewed, perhaps moreso than if you were composing it yourself... And the additional variables include things like finding out of the box solutions either with or without the model outputs, which is hard to control for as well. Even more impossible to control variables are model quality, training material related to the topic, and many more of those classes of issues that may not be publically available or even possible to trace fully. Truly, anecdotes are not informative of the general experience. It is literally the cliche that is mentioned on this site periodically "these are the lotto numbers that worked for me."

For me it feels like roughly a 10–20x change, but mostly because I restructured how I work rather than just adding an “AI helper” on top.

In the last year I’ve shipped a couple of small OSS tools that I almost certainly would not have finished without AI‑assisted “vibe coding”. Everything I build now flows through AI, but in a slightly different way than just chatting with an LLM. I rarely use standalone ChatGPT/Gemini/Claude; almost all of it happens inside GitHub with Copilot and agents wired into my repos.

The big shift was treating GitHub as the interface for almost all of my work, not just code. I have repos for things like hiring, application review, financial reviews, and other operational workflows. There are “master” repos with folders and sub‑folders, and each folder has specific context plus instructions that the AI agent should follow when operating in that scope, essentially global rules and sub‑rules for each area of work.

Because of that structure, AI speeds up more than just the typing of code. Idea → spec → implementation → iteration all compress into much tighter loops. Tasks that would have taken me weeks end‑to‑end are now usually a couple of days, sometimes hours. Subjectively that’s where the 10–20x feeling comes from, even though it’s hard to measure precisely.

On the team side we’ve largely replaced traditional stand‑ups with AI‑mediated updates. KPIs and goals live in these repos, and progress/achievements are logged and summarized via AI, which makes updates more quantitative and easier to search back through. It’s starting to feel less like “AI helps me with code” and more like “AI is the main operating system for how our team works.”

Happy to share more about the repo/folder structure or what has/hasn’t worked if anyone’s curious.

I’m not sure you’re going to get any useful results from this question.

Some people find it useful, some people don’t, and unless what you’re using it for matches what they’re using it for (which you’re not asking) none of the results you get give you any insight into what you should expect for your use case.

Oh well, whatever. Here’s my $0.02; on a large code base that takes up to 30 minutes to do a local type check in typescript, the net benefit of AI is neutral or negative, because the agent can’t loop effectively and check it’s own results.

AI scaffolded results are largely irrelevant and don’t use internal design system components or tokens for UI and are generally useless.

Objectively measured ticket completion rates are not meaningfully impacted by the use of AI.

Out of date documentation leads agents to build incorrect solutions using outdated and depreciated techniques and services.

This is true across multiple tools and multiple models used, including sota.

1x

It is not more productive.

This reflects on my personal experience in the last 8 months of intense (and company mandated) AI usage at work.

At home, for small personal projects, I would say it’s closer to the 2x you describe, maybe as much as 3x for building and iterating on rich web UI using react.

I'd be interested in how you setup those repos for non-coding tasks, thanks for sharing!