I’m working in Triple Minds and I’m working on a Candy AI Clone project like this. Everything’s working fine – chat, memory, and character logic – but I’m struggling with the typing delay simulation. I want the AI to look like it’s “typing” naturally before sending messages, but the timing feels off. Sometimes it types too fast, other times way too slow 😅
Here’s what I’ve done so far:
Added a dynamic delay based on message length
Used frontend timeouts tied to token count
Tried async streaming from backend for smoother delivery
Still, it doesn’t feel human enough or consistent
Has anyone here implemented realistic AI typing animations or live streaming chats in their Candy AI clones? Would love to hear what delay logic or UX pattern you used.
Comment