I believe you should do what you genuinely find interesting. Go for 1, dig into internals, read some papers, and see how it goes. Even if you decide not to get into ML/AI, learning how stuff works is always rewarding.
The audio on this is about the worst I’ve ever heard on YouTube. I fast forwarded and at least he stops playing that loud music over his quiet voice, but damn.
He gets off topic a lot (bullies, amphetamine salts??) and spends the entire time talking to the commenters not the video recording.
Surely, there’s a better video out there than this.
I think it's more the hand-in-handedness that seems to exist between "rewrite an existing, mature tool" and doing it in Rust. Half the time it's hard for me to know which caused which — the need for the tool, or the desire to rewrite something in Rust.
The other options are C, C++, Go, and maybe Ada or Zig, though I haven't seen many CLI tools written in those two in practice. In practice, it seems like Go, Rust, and C++ are the preferred languages for newer CLI tools, although I have no data; my conclusion is based on my general perception. Older ones, C and Perl.
I'm a lot happier with a fad for Rust-written CLI tools than the disappointment of reading install instructions for a simple CLI tool that starts with "First... npm... bower..."
V avoids doing unnecessary allocations in the first place by using value types, string buffers, promoting a simple abstraction-free code style.
Most objects (~90-100%) are freed by V's autofree engine: the compiler inserts necessary free calls automatically during compilation. Remaining small percentage of objects is freed via reference counting.
The developer doesn't need to change anything in their code. "It just works", like in Python, Go, or Java, except there's no heavy GC tracing everything or expensive RC for each object.
