Make it possible to set up a personal filter of words. Sentences using that word will automatically no be displayed for me no matter what channel Im in.

Why ?
Many streams, especially of large streamers are filled with spam that provides no value for me and makes the chat unreadable. Examples are pepelaugh, kekw, lulw, pog etc.
This is a personal preference and I dont think its realistic that I should need to run around and convince each streamer to ban the words I dont like.
I also tried blocking all users using those words but they are so common so even after spending 30 minutes of blocking people the chat is still full of these words.

This should have the additional benefit of saving bandwidth for twitch.

How about:
1. Set up a system to analyze users reported for being suspicious of being a bot in a chat.
2. Said system allows the person reporting to select one or more recent chat messages from said user.
3. The chat message gets added to a staff-curated list of spam messages. The curated list runs a check behind the scenes (maybe via a JS that already does the "sending message" mechanics).
4. If the message matches the list, the account is flagged for investigation and the message is shadow-banned (the message appears for the sender, but anybody else won't see the message because it wasn't actually sent)

Currently, moderator status is unreliable. This idea is to have a reliable way to know if I am a moderator¹.

Please provide a reliable message of changes to moderator status ASAP to the affected user. Something along the lines of the standard IRC MODE message—but that isn't affected by the MODE spam tmi tended to generate.

Additionally, have a short grace period (i.e., when excess messages are discarded without resulting in a ban) after moderator status is revoked to account for latency in receiving the updated status.

1. Bots currently have no reliable way to know if they are moderators (except in their own chats), so they must follow the same limits for non-moderators. If bots relied on the currently unreliable status, this would potentially open an attack vector to cause the bot to be banned. This suggestion is to correct that oversight, and account for the delay in the status update being received.

Many Twitch channels are essentially podcasts, auditory ASMR, and otherwise are just fun to listen to. Sometimes people will also have multiple streams open in the event of a collaboration and be regularly switching back and forth.

It doesn't make sense to only make a view count on the active tab when people are bound to multitask. I'm sure it is because of advertisers, but ads have audio cues as well. Not to mention, a lot of times the people who will inevitably be multitasking (because who can sit still for a 6+ hour stream) are subs who have shown their support for the channel.

It feels disingenuous, and the viewer count takes so long to update that I find that for both my channel and channels I mod in, the number of people actively chatting and engaging with the streamer (so i know the stream isn't muted either) will often be higher than what the viewer count displays.

To me, it makes more sense to have the viewer count align more closely with how many people are in chat, but then things like 'active tabs' should be analytics available to twitch and the streamer on the back end if they so choose.

Not to mention, many potential stream participants might be turned off from a stream if the view number seems too low or the community seems too small when in reality people are probably tabbing in and out to participate.