Comment by crystal_revenge
Comment by crystal_revenge 3 days ago
For me (and most of my friends/coworkers) the point of AoC was to write in some language that you always wanted to learn but never had the chance. The AoC problems tend to be excellent material for a crash course in a new PL because they cover a range of common programming tasks.
Historically good candidates are:
- Rust (despite it's popularity, I know a lot of devs who haven't had time to play with it).
- Haskell (though today I'd try Lean4)
- Racket/Common Lisp/Other scheme lisp you haven't tried
- Erlang/Elixir (probably my choice this year)
- Prolog
Especially for those langs that people typically dabble in but never get a change to write non-trivial software in (Haskell, Prolog, Racket) AoC is fantastic for really getting a feel for the language.
Yes, this year I'm going for Lean 4: https://github.com/ngrislain/lean-adventofcode-2025
It's a great language. It's dependent-types / theorem-proving-oriented type-system combined with AI assistants makes it the language of the future IMO.