The core Android team is four engineers so about double in size.

This makes some sense since on Android we have to re-write most of the stores/business logic and not just UI. We have modeled the architecture similarly to things on the Desktop/iOS side so we still are able to move fast and stay lean as a team.

Can you expand/clarify on what you mean by "on Android we have to re-write most of the stores/business logic and not just UI"?

The iOS application can safely pull data and perform actions from just about every Store (we don't use redux but the idea is the same https://redux.js.org/api-reference/store) and Action Handler (https://reactjs.org/docs/handling-events.html) that exists in the main Discord React application.

So if one is writing a feature that say allows one to ban a user - they don't need to write any of the banning logic or worry about fetching the user data/validating it against server roles permissions - if it exists on the main Discord app, it can be safely imported.

Other helpful examples are perhaps things like markdown parsing - iOS was able to just import and use the exact same system desktop uses to handle markdown and things like user/channel mentions, custom emoji, etc. On Android we had to write one ourselves: https://blog.discordapp.com/how-discord-renders-rich-message...

On Android since there is no code re-use everything had to be written from scratch.

Oh, I missed that the Android version isn't React Native.