cocoto

I’m pretty sure it’s for security reasons, it allows the app to ensure the code comes from a correct Signal phone number and not from a third party actor. It’s hand holding yes but it’s a trade off in favor of non-technical users.

Yes the default behavior of working around the bug is good, that’s what everyone was doing anyway pre-LLMs. The good practice is to report the bug. Simply add instructions to your agents to report you the bugs it found in the summary, so you can check the bug and report it correctly upstream.

Except for doing maths, which is painful to input on screens (but beautiful).

It also helps that your history with the doctor will help you build trust. If you have the same doctor for years and you only went twice for symptoms (that was treated by specialists), then your doctor might redirect you to a specialist in no time as soon as you have any new symptoms.

Backup codes, sms backup, trusted user backup. Most of these are quite insecure but better than loosing access to your digital life.

> The du options are: s to summarize and b to display bytes.

Small nitpick: Use long options and your code samples become self explanatory!

Too pricey for food, it’s cheaper to print.

You can edit single files in most git forges and it will open a pull request for you, the workflow is not that bad.

This doesn’t scale at all for some problems where a specific algorithm is needed (graph problems for instance).

“do” imply “can”, but “can” do not imply “do”.

I recognize myself in the first paragraph (except being totally exhausted to the point having to sleep). And I started working as a SWE 4 years ago (so full dopamine brainrot access). Maybe it is because I have a reasonable usage of LLMs?

You forgot the “autonomous” part of the definition.

I have a near total opposite view as a cyclist in France. Buses are extremely dangerous in cities, and are way louder than cars.

If AI coding agents follow the same path then any software engineer over 25yo can peacefully continue working for 30 years until the tech is capable. The comparison is futile…

N = 1. The company I’m working for, (a hundred employees, EU), we basically build projects on data science, ML, optimization, operational research and such for many industries (transportation, manufacturing, finance, engineering, etc). We are basically the first to know about the latest advances in AI and we have internal optional workshops on LLM assisted development. And yet nobody is pressured to use these tools. At most we are incentivized to at least try these tools.