They're pretty similar, but with different ergonomics. Algebraic effects are similar to some kind of "free" monad technique, but built in. For being built in they have nicer syntax and better composability, often. You can achieve the same in a language suitably dedicated to monadic approaches (Haskell being the poster child here) but it helps to have type class inference (giving you mtl-like composability) and built-in bind syntax a la Haskell's `do` or Scala's `for`.