Some questions on my mind. I'd love to chat if you're also interested in these things.
If not microservices, then what?
If microservices are an anti-pattern, what should we do instead?
My hunch is that cloud software development will evolve similar to how desktop software developement evolved as compiler technology improved.
Maybe we'll revert to monoliths, but with more infrastructure automation and compiled glue logic.
Why is symbolic ML not more popular?
Is it just hard?
I'm skeptical that current black-box deep learning methods will get us where we need to be. They're too fragile.
If an AI system can't explain itself to a 5 year old, with provable guarantees that the explanation is complete, I'm not interested.
What would it take to build an algorithm which can learn how to emulate Conway's Game of Life, and prove it by explaining the rules in terms I understand?
How can we reduce 3rd-world country investment risk?
I'd love to invest time and money into my birth country, Zimbabwe. But I don't trust the government or the infrastructure enough to risk it.
Solutions might involve crypto-currencies.