- Unreliable and buggy core libraries (basic fp math, stats, autodiff) which seems to stem from an academic-style disinterest in focusing on the boring bits of

- Unreliable and buggy core libraries (basic fp math, stats, autodiff) which seems to stem from an academic-style disinterest in focusing on the boring bits of language foundations - 1 indexed arrays - extremely slow start up times (1 min??) - no way to create portable static binary apps (disadvantage vs rust/go) - lack of tutorials, demos - julia is not really faster for math when compared to jax/tf/pytorch/etc anyways