For example, Latin language and Roman numerals. Are people likely to mistake numbers like ID (499) or VIM (994) for words, or is it always clear enough what's what from context? I think Hebrew, Braille, and maybe Greek also do this, though i'm not as familiar with those scripts as with Latin.

I ask because i'm making a conlang and having a little trouble coming up with enough letters, let alone numerals too. Reusing letters would be helpful and probably not confusing if i make sure that numbers are never pronouncable as letters, but making this easy to read is important to me.