NobBot.tv

Music can be a big part of a smaller Twitch stream’s identity. It gives viewers something to talk about, creates an easy way to participate, and helps a channel feel like its own community.

Existing song-request tools handle the basic interaction, but their queues often become unfair or repetitive. One viewer can submit several songs at once, the same tracks and artists keep appearing, and streamers must either manage everything manually or introduce strict limits.

There is also a larger discovery problem. Twitch primarily organizes streams by game or category. Viewers have no practical way to find smaller communities based on the music they play or the tastes of the people watching them.

NobBot started as a better way to manage requests, with the longer-term goal of helping viewers discover Twitch communities through shared music tastes.

NobBot rotates between viewers instead of playing every song in the order it was submitted. People can request multiple tracks without one person filling the playlist or preventing everyone else from participating.

Streamers can also set cooldowns for recently played songs and artists. This reduces repetition while letting each channel choose rules that fit its own community.

The rules are visible to the person making the request. When NobBot rejects a request, it explains why. Accepted requests show their position in the queue.

Processing a request was not enough. People needed to understand why it was accepted, rejected, or moved and trust that everyone was getting a turn.

Streamers already have a lot competing for their attention during a broadcast: the game, chat, broadcasting software, and the audience itself.

NobBot brings the main music controls into one interface, including:

• The current track and playback controls
• The live queue and requester information
• Queue moderation and reordering
• Channel-specific request settings
• Listening history and community statistics

Viewers have their own public channel page where they can check the current track, follow the queue, and explore recent activity without repeatedly asking for updates in chat.

Both sides stay connected. A request submitted through Twitch chat appears in the dashboard, and changes made by the streamer are immediately reflected in the public queue.

Every request says something about the channel: which songs keep returning, which artists connect with viewers, and who is helping shape the playlist.

NobBot records listening trends, popular music, and viewer participation. During the alpha, viewers responded most strongly to the listening statistics and requester charts. They compared the results and talked more about the music being played.

As NobBot expands to more channels, the same activity could help viewers find streams that match their tastes. Instead of browsing only by game or viewer count, someone could discover a smaller community through familiar songs and artists.

The alpha tested the queue and channel-level analytics. Using that data for discovery across multiple channels is the next stage of the product, not something the initial launch has already proven.

NobBot’s queue can change from several places at once. Viewers add songs through Twitch chat, while streamers can skip, remove, promote, or reorder requests from the dashboard and player. Every change still has to preserve the rotation between viewers.

I built the queue around shared state in Firebase and used transactions when a request and the current queue position needed to change together. If that state is missing or invalid, NobBot rebuilds it from the individual requests and restores whose turn should come next.

The player, manage screen, and public queue subscribe to the same live state. A song added through chat or skipped in another window appears without refreshing. Reconnect handling keeps channels from losing or duplicating their Twitch connection, while rate limits prevent overlapping commands from overwhelming the service.

NobBot ran successfully during live Twitch streams. Streamers managed requests from the dashboard instead of relying only on chat commands. Viewers responded most strongly to the listening statistics and requester charts, which gave chat another reason to talk about the music.

The next release is planned for more than 10 active streamers. It will test onboarding and day-to-day reliability across different channels. Using listening data for cross-channel discovery remains a later stage of the product.