The strongest competitors or a government law? I prefer the second.

Suppose Chinese model are super-risky: is it better to have them in the open, so everyone can review the problems or have a closed-model?

Security is better at open Linux is open source because development is better.

I understand Anthropic for sure, but trying to limit open weight model seems the opposite direction IF security+safety is the major concert.

The Genie is ALREADY outside the bottle, and there are already plenty of companies offering inference services

>Right now defenders are effectively banned from using Fable or Sol for cybersecurity because of Trump administration directives; that means the best alternative is using models from a country which has been trying to weaken our cyber defenses for years. This is insane!

I understand some guardrails are needed, but it is becoming increasing problematic manage them without a strong public discussion.

Comments URL: https://news.ycombinator.com/item?id=48932178

Points: 2

# Comments: 5

I mean before asking AI to implement :)

I built a template repository to govern your AI chats with more grip than usual. It leverages PI.dev coding agent and an aggressive isolation done with Docker + VS Code

So far works on Linux and MacOS, and should also works on WSL2.

The project uses a container to run your preferred AI harnesses CLI. The container is optional, but protect your disk from bad things (I noticed rm -rf is always the preferred removal command of Deepseek :) a bit scary)

I am slowly extending it, aiming at

- reduce token usage - ability to install extra skill via pi.dev - experimenting with variants (like AWS AI-SDLC v1).

Please feel free to comment and give me hints.

Comments URL: https://news.ycombinator.com/item?id=48894720

Points: 1

# Comments: 0

In all my professional carrier so far, a full rewrite without reason is a recipe for disaster, see

https://www.joelonsoftware.com/2000/04/06/things-you-should-...

This Joel Sposky article is always my northstar when I have such "temptations". My 2 cents

https://github.com/daitangio/take-ai-control

It is docker + vscode friendly. I tested it with major systems (copilot, codex, Claude Code and pi.dev) Comments Wellcome!

https://github.com/daitangio/misterio

It works very well!

SQLite is an embedded database: no socket to open, you directly access to it via file system.

If you do not plan to use BigData with high number of writers, you will have an hard time beating SQLite on modern hardware, on average use cases.

I have written a super simple search engine [1] using python asyncio and SQLite is not the bottleneck so far.

If you are hitting the SQLite limit, I have an happy news: PostgreSQL upgrade will be enough for a lot of use cases [2]: you can use it to play with a schemaless mongo-like database, a simple queue system [3] or a search engine with stemming. After a while you can decide if you need a specialized component (i.e. Kafka, Elastic Search, etc) for one of your services.

[1]: https://github.com/daitangio/find

[2]: https://gioorgi.com/2025/postgres-all/

[3]: https://github.com/daitangio/pque

[4]: https://docs.paperless-ngx.com

It is quite easy to setup and give you much control, I do it myself:

https://gioorgi.com/2020/mail-server-on-docker

Documentation is very very clear

The reason is the anti-spam rules and the fact that Google, Microsoft and so on are creating a iron trust to each other, and the little server outside are marked spam by default.

Lets encrypt avoided a similar destiny to https connections, but the risk is always out of the window. I mean, https was becoming "pay-us-to-publish a web server, or our browser will mark you as unsafe and do not display it".

I think it is time also to self-host private free chats and possibly other services lik DDoS services.

I am GMail-free from 2020 and I am a lot happier.

I keep my GMail account, but I use just as failsafe or as seconday account. In the meantime, I have very little trouble managing my mail-server.