- https://pyinfra.com - Python infrastructure management, been picking up pace recently trying to keep on top of PRs!
- https://kanmail.io - always plugging away at Kanmail changes, no one uses it really except me but that’s fine :) new site designed by fable blew my mind
- https://verified.fyi - first side project started initially as a vibe coding experiment, but it’s become quite good at sussing out dodgy websites
Indeed! (I am that original contributor :)), lots of work ongoing to address this, we now have a small maintainers group and are sharing out review and release loads.
This! Been trying to find the best (least worst) solution to this since 2015 when I started pyinfra. Done ast parsing/hacking, done weird context managers instead, tried rewriting statements to context managers. _if is the latest, and I think least worst, option right now.
Basically a flaw of the entire model where you write code as if executing a single host which is then executed on many in parallel, forcing the two step diff and deploy that causes this.
Funny thing is since v3 this behavior (diff then execute) is even desired with the yes prompt like terraform.
Absolutely zero sympathy. You’re responsible for anything an agent you instructed does. Allowing it to run independently is on you (and all the others doing exactly this). This is only going to become more and more common.
One silver lining this is finally going to push me to switch to a dedicated camera and some niche unrestricted Linux or graphene device as a phone. Goodbye iPhone. (I say this as someone with an Apple account old enough to auto “qualify”, how lucky).
Also parent in the UK - strong disagree, it’s part of our parental responsibilities to set this up, not doing it is the same as not watching a newly walking baby on the stairs (/etc). Compromising everyone’s privacy for a subset of lazy parents is a failing of society.
All of these agent wrappers/claws/whatever seem to encourage wildly irresponsible usage of AI. Sure it can get stuff done in what is effectively a simple loop but without review we’re just playing with fire.
I was experimenting with Claude code doing similar (don’t need wrappers really) and the result was a huge amount of mediocre code and my weekly limit burnt. The code “works” but oh my the duplication and potential bug surface is off the chart.
Stopped my experiment and back to human in the loop plan->execute cycle which is way more effective.
Thankfully I don’t think we’re at jarvis levels yet.
As a primarily Go dev - 100% agree. The endless check and wrap error results in long chains of messages you have to grep for to understand the call stack. For what benefit? Might as well just panic and recover/log the stack in many cases.
The key here is a singleton sequencer component that stamps the new versions. There was a great article shared here on similar techniques used in trading order books (https://news.ycombinator.com/item?id=46192181).
Agree this is the best solution, I’d rather have a tiny failover period than risk serialization issues. Working with FDB has been such a joy because it’s serializable it takes away an entire class of error to consider, leading to simpler implementation.
Eight, but half of them exist as long running test accounts for Kanmail (kanmail.io) so arguably do not count. I do use them for legit emails though, new services usually get one of four test emails.
- https://kanmail.io - always plugging away at Kanmail changes, no one uses it really except me but that’s fine :) new site designed by fable blew my mind
- https://verified.fyi - first side project started initially as a vibe coding experiment, but it’s become quite good at sussing out dodgy websites