Aerialoo

AI generates or not - irrelevant if the substance is high quality. Always good to see projects. You should consider hosting this leaderboard on hugging face, you will get a lot more AI enthusiasts and practitioners there.

I seriously don't understand the apple glazing. They are pretty much completely out of the picture and are one of the most anti consumer anti open-source companies. Yet developers STILL hope for Apple acquisitions and being saved... Or something like that... We have llama.cpp, vLLM, opencode, hermes and countless other mature great open source tools that don't require a anti-consumer company's hands in the jar.

I think this experience is universal. The answer is the same as always has been - develop skills that are becoming most important. Right now that is (at least from what I can see): - Data analysis, data pipelines, models, etc. - Tacit business knowledge - architecture and design patterns (always has been, but now the scale os larger so this is even more important)

It's harsh but nobody cares if a model or a human made a system.

The "good" bits are that now automating anything and providing value from software is much easier. If I have an idea or a nitpick somewhere, I can just do it, up to a limit (which is quickly rising).

I have always been a generalist and generally interested in a very wide array of things, and this period has been the most exciting in my engineering career (13y now). Learning about anything is so frictionless, looking back at my first learning experience - picking up a fat C++ book and spending days/weeks debugging, while I can romanticize that, I would never go back.

I can also now write software solo or with an extremely small team at a huge scale in comparison, and that is super exciting.

A lot of skills that took sleepless nights to acquire, they are "gone", but I still don't regret anything or wouldn't go back. Their "usefulness" has degraded, true, but this has always been the case with engineering.

We are now able to spend much more time thinking about utility rather than low level implementation and imo that's great.

We have many challenges ahead of us, and there are seriously bad things, the biggest one I have experienced is the hours are increasing and mental load is vastly increasing as well. As capacity, speed and leverage increases, so do expectations and hours, and that is probably a social problem.

Sorry for the unstructured stream of thoughts, and this is just an opinion (quite an unpopular one I believe), I hope your distress decays away for a new excitement and new opportunities.

Thanks for the article .

Use a bigger fan

Your music taste is similar to mine, here's some stuff you might like:

nthng Skee Mask vc-118A Mohlao - landforms Traumprinz/Prince of Denmark/Prime minister of doom/etc. (Same guy) Giulio Aldinucci - Shards of distant times (ambient) Wanderwelle (ambient) Quiet places (the whole label A strangely isolated place is amazing, also ambient) Bowery Electric (also Boards of Canada from the 90s) Belong - October Language Rafael Anton Irisarri

I explore a lot of new electronic music and the artists you mentioned I was obsessed with a year ago or so. Since then, I have been digging deeper and deeper, enjoy.

CoinJoins exist exactly to fix the problem of fungibility.

I can assure you that's not true, a lot of good developers like Chris Belcher are actively working towards it and even receive grants. New BIPs also propose privacy improvements. Anyone can care about privacy. It's a pseudo-anonymous network, but it can be used anonymously.

The vigilantes? They are literally trying to destroy privacy and fungibility on the network. Thank god for CoinJoins.