ddaniel10

Yes.

My point in this experiment, is that we as humans all we need to do is to give to all ai agetns one data source to write,read,edit and delete, and they will be able to handle everything for us. this is the point from my research that i'm most inspired.

If tomorrow all agents are aligned to the same list/data source, the current infrastructure of all apps and software will be changed, because we won't need it anymore

i dont get it. independent shops you mean airbnb?

its really interesting question about "independent shops" as separated topic, i guess they actually going to benefit from this no?

Nice. what the use cases and groups of people you see that are going to use it?

I really liked to read this. i had this idea in mind but its nice to be remind of it

Hi, can you please contact me at [email protected]

Thanks for the feedback. Are you going to forget this name though?

Haha it's your personal agent, let him handle the stuff you don't like. But soon, right now its not fully ready

Great point. I wrote it as important note and ill take it into account.

I mean, you can just say Hi to it, and it will cost nothing. It only adds code and features if you ask it to

DIYWA - do it yourself with agent ;) hopefully zuckerman as the start point

I'm building this in the hope that AI will be cheap one day. For now, I'll add many optimizations

thanks

Hi HN,

I'm building Zuckerman: a personal AI agent that starts ultra-minimal and can improve itself in real time by editing its own files (code + configuration). Agents can also share useful discoveries and improvements with each other.

Repo: https://github.com/zuckermanai/zuckerman

The motivation is to build something dead-simple and approachable, in contrast to projects like OpenClaw, which is extremely powerful but has grown complex: heavier setup, a large codebase, skill ecosystems, and ongoing security discussions.

Zuckerman flips that:

1. Starts with almost nothing (core essentials only).

2. Behavior/tools/prompts live in plain text files.

3. The agent can rewrite its own configuration and code.

4. Changes hot-reload instantly (save -> reload).

5. Agents can share improvements with others.

6. Multi-channel support (Discord/Slack/Telegram/web/voice, etc).

Security note: self-edit access is obviously high-risk by design, but basic controls are built in (policy sandboxing, auth, secret management).

Tech stack: TypeScript, Electron desktop app + WebSocket gateway, pnpm + Vite/Turbo.

Quickstart is literally:

  pnpm install && pnpm run dev

It's very early/WIP, but the self-editing loop already works in basic scenarios and is surprisingly addictive to play with.

Would love feedback from folks who have built agent systems or thought about safe self-modification.

Thanks :)

Hi, thanks for sharing

We have multiple data sources, not just scraping. There are many solutions for this, but they can be expensive. For now, we’re using some infrastructure that we built ourselves, and it’s enough for our small scale.

Thanks for sharing :) So as I wrote "Scholtz is currently focused on LinkedIn profiles in the U.S. tech industry."

Thanks for sharing. That’s interesting. I don’t think it’s an issue with AI itself, but rather with the data. I’m almost certain—99.999% sure—that if I had provided it with the right datasource, it would have fetched it.

Also, the search is open for editing through memory and context dialogs. So in the future, you’ll be able to write, “I need one with many Google reviews.”

I think one of the issues models has today that makes people worry when consuming information there about history and conflicts that it has mistakes.

If you can solve this issue, it would be great. Show grounded texts: Left/Blue side says: Right/Red side says:

Because history is not math so each group has its own history