cd /news/ai-tools/ask-hn-i-think-i-accidentally-built-… · home topics ai-tools article
[ARTICLE · art-31598] src=news.ycombinator.com ↗ pub= topic=ai-tools verified=true sentiment=· neutral

Ask HN: I think I accidentally built a better cloud agent solution. Hubris?

A developer built Else, a tool initially designed for product managers to prototype on top of their actual product, but now uses it as a personal cloud agent development environment. The tool differentiates itself by focusing on the app rather than the code, running on real URLs, and allowing agents to create overview pages with videos and state-loading buttons. The developer is considering open-sourcing the local version as an alternative to existing tools like Conductor and Superset.

read2 min views1 publishedJun 17, 2026

We built Else, a tool for PMs to prototype on top of their actual product.

I started using Else to build Else, and I now inadvertently spend all my time turning it into my ideal dev environment.

Else works similar to most cloud agent systems with some key differences that I personally think have a huge impact, but I’m worried I have tunnel vision.

WHY DO I THINK IT'S SO GOOD?

It’s equally good at the first 80% (the agent spends an hour building the feature) and the last 20% (you need to test, iterate, fix issues, and polish the UX).

  - It’s built around the app, not the code. Think Lovable, v0, etc. The app runs on a real url (no remoting into localhost).
  - When building a feature, the agent also creates an overview page with videos and buttons that load the correct db and frontend state. Saves tons of time when you manually verify.
  - If a tool or state snapshot is universally useful, save it. Over time, you build an extensive development dashboard.
  - Else can also run locally using Docker containers instead of VMs.
  - Else can edit Else. If you want something to work differently in the dev tool itself, just tell the agent to change it.

WHY AM I WORRIED IT'S HUBRIS?Isn’t this what Devin, Cursor cloud agents, etc, etc, etc are trying to solve? These are my pain points, but maybe I'm just not using them right?

  - Getting the full stack running is a major PITA, and it’s just as much of a PITA to maintain
  - I personally hate the “remote desktop into a browser so you can use localhost” approach. It’s clunky and limiting, and it makes the last 20% painful. I usually end up switching back to local development, which has its own issues.
  - They feel rigid. I want to tweak things to my preferences, but I don’t want to spend time figuring out how.

What about working locally with Conductor, Superset, or just Claude/Codex?

  - I don’t want to live in the terminal.
  - I still need to figure out how to run five instances of my app concurrently.
  - Once I’ve solved that, it’s still a headache to keep track of which agent is on which port.

SO TELL ME HONESTLY:

  - Am I high on my own supply? Have I just not spent enough time figuring out the best way to use the tools already on the market?
  - I’m considering open-sourcing and releasing the local version as a Conductor/Superset alternative for people who don’t want to live in the terminal. Would you be interested in trying it?

Comments URL: https://news.ycombinator.com/item?id=48574681

Points: 1

── more in #ai-tools 4 stories · sorted by recency
── more on @else 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/ask-hn-i-think-i-acc…] indexed:0 read:2min 2026-06-17 ·