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

Ask HN: What is your (AI) dev tech stack / workflow?

news.ycombinator.com

Post reply on HN
Print view View original

141–150 of 187 posts

Re: Ask HN: What is your (AI) dev tech stack / workflow?

#141

Post by jpeeler »

I'm currently using herde[1] to handle/supervise multiple agents (with some patches I need to try to upstream) along with Nono[2] for sandboxing. This sandboxing approach avoids use of a microVM, which lets me use tooling I already have installed inside the sandbox. The downside is getting all the policies correct as it seems every project needs some new type of access, though Nono does try to make policy writing easy.

I've been considering switching my approach to using a microVM through microsandbox[3]. The pro of this approach is you can essentially skip the policies and rely on the security of the VM boundary. The negative is that now you've lost all your installed tools, so you need to either provision at runtime or build something (like an image) beforehand to match your dev environment.

I still don't know which is less maintenance. And while I think herde is pretty well thought out, I do think about something that works outside the terminal may be nicer.

[1] https://github.com/ogulcancelik/herdr

[2] https://github.com/always-further/nono

[3] https://github.com/superradcompany/microsandbox

Re: Ask HN: What is your (AI) dev tech stack / workflow?

#142

Post by sermakarevich »

madduci wrote: post #127
sermakarevich wrote: post #54

Earlier quoted context omitted.

I am building AI agents full time since Nov 2024. I stopped coding completely around mid summer 2025 using Cursor at that time. When you build platform-like application, and have few plugins already, ai coder can create next one in a way you won't recognize which one is written by you. At the end of 2025 I switched to Claude Code. Compared to Cursor this opened a different level of automation, including fe possibilit…

What about quality? How do you assure that everything works without misshapen?

TDD and specs help

Re: Ask HN: What is your (AI) dev tech stack / workflow?

#145

Post by amdolan »

jpeeler wrote: post #141

I'm currently using herde[1] to handle/supervise multiple agents (with some patches I need to try to upstream) along with Nono[2] for sandboxing. This sandboxing approach avoids use of a microVM, which lets me use tooling I already have installed inside the sandbox. The downside is getting all the policies correct as it seems every project needs some new type of access, though Nono does try to make policy writing eas…

Thanks for sharing. I recently started using my homemade “competitor” to herde, so it’s nice to compare against prior art.

What do you think could be nicer with a native app? More mouse or visual interactions? Modern design and gui?

Re: Ask HN: What is your (AI) dev tech stack / workflow?

#146

Post by abelkhadir »

I use linux + claude code as the main driver.

Before i start a project, i first write scripts manually to give claude code an idea of what im trying to do

I then use Claude design to design the full frontend in one shot. then use it manually to see bugs/stuff i dont like and re-prompt it to modify. i also give it description of the scripts i wrote so it has a better idea of what i want

Then i use claude code for the backend. depends on task, i mostly use python, golang or typescript.

No IDE. Terminal only. git + editor + agent is enough.

i used this to code my entire project kyroai.dev. it turned out impressive, in under a week

I also belive you should be the number one user of your app to actually make it bug free

Before this i had to type code similar to you, and i say ai code is way faster, for example kyro would have taken me a month for the mvp. but with ai i just prompt. while its working, i browse the app, find newer bugs/changes re prompt again, in under a week the project is done.

All of this with a 20$ claude sub. not even a 100$ plan

Re: Ask HN: What is your (AI) dev tech stack / workflow?

#149

Post by cadamsdotcom »

The right set of techniques is highly situational and HN is full of people just dipping their toes into agentic coding.

I’ve been using Claude Code for over a year and built a 300k+ saas codebase with it.

I use zero skills, zero MCPs, zero adornments to the base model.

I have built something very different from what others have: my “stack” is a collection of Claude Code hooks that force the model to do test driven development, also known as TDD or “red/green/refactor”. It’s a process for building software that’s decades old, but was always a PITA to do solo. It’s kind of like the scene in karate kid - wax on, wax off…

- Write failing tests. - Watch them fail (if they pass, whoops! Your tests are wrong!) - Write code that makes the tests pass - Witness them pass - Optionally refactor (since you have tests that you saw fail and made pass - you can refactor with impunity) - Repeat from start..

The agent is forced through this loop. The automation is deterministic and uses a state machine. Automation forces the model write end to end tests and unit/integration/component tests before it’s permitted to write code. Witnessing tests fail first proves that they test something, and depending on e2e tests means that in a year of building I’ve never had to hand my model a browser. Instead it will write Playwright code to validate its work, and that becomes forever part of the codebase, preventing regressions forever.

Even despite this, I test everything manually and review every line I merge.

It takes 10x as long to get a change into shape for merging as the model spent building it. Even though it emerged in a working state from the model no lint errors.

Working at this level of complexity for 40-50 hours a week, you end up glad for the weekend - or glad to do prototypes. When you can do stuff that’s one-shottable you absolutely should.

When I want something as simple as you’re describing, I write prompts and test by hand. I rarely if ever read the code.

In other words, it’s highly situational.

If you want to see my process, the codebase is at https://codeleash.dev.

Thanks for reading!

Re: Ask HN: What is your (AI) dev tech stack / workflow?

#150

Post by jpeeler »

amdolan wrote: post #145
jpeeler wrote: post #141

I'm currently using herde[1] to handle/supervise multiple agents (with some patches I need to try to upstream) along with Nono[2] for sandboxing. This sandboxing approach avoids use of a microVM, which lets me use tooling I already have installed inside the sandbox. The downside is getting all the policies correct as it seems every project needs some new type of access, though Nono does try to make policy writing eas…

Thanks for sharing. I recently started using my homemade “competitor” to herde, so it’s nice to compare against prior art. What do you think could be nicer with a native app? More mouse or visual interactions? Modern design and gui?

The kitty developer is philosophically against multiplexers, but he also has concern about performance (which I share somewhat as well). Herde is not quite as fully featured as zellij/tmux, so that partially is biasing me. I think that a native app could be potentially nicer to avoid all those concerns, along with my setup (only in Linux really) of a tiling manager handling the windowing instead of that being part of the solution. A big part of the value add of Herde is actually the monitoring of the Claude session and not so much the multiplexing reimplementation. One of my patches allows jumping to the next blocked/finished session so I can quickly give feedback or observe different agents. I can envision a GUI solution for that workflow being something better than what a terminal easily allows.
Post reply on HN