Inspiration
As proud doomscrollers, we often find short-form clips from big streamers on our "For You" pages. But these streamers aren’t the ones posting the content—they're too busy managing the stream itself. Instead, random individuals repost clips shortly after the stream and monetize them. This results in lost revenue and visibility for the streamer.
What it does
Our project solves this problem while offering additional benefits. It automates the clipping, editing, and posting process for the streamer in real time, allowing them to regain control over their content. By monitoring audio, video, and chat activity, it identifies moments worth clipping. When triggered, the system generates, formats, and posts the clip across the streamer’s social media platforms—automatically and instantly.
This background process is faster than any human, effectively outpacing manual reuploaders. Since it can publish clips while the stream is still live, it also acts as a form of cross-platform promotion, redirecting traffic to the ongoing stream and boosting viewership.
How we built it
At the core of the project is an RTMP server written in JavaScript, which receives a duplicate live feed. This feed is stored in a rolling buffer and analyzed in real time. A Python backend, hosting an AI model trained to detect facial expressions, processes the video, audio, and chat data to decide whether a segment should be clipped.
If a moment is flagged, the backend formats the content for vertical screens, stores it in a cloud database, and publishes it on the selected platforms. A React TypeScript web application allows streamers to authenticate, connect to the RTMP server, and view data on the clips posted during their livestream.
Challenges we ran into
One of our initial goals was to be platform-agnostic—both in terms of capturing the stream and posting the content. Achieving this was more difficult than expected, especially given the variability in APIs, media formats, and content policies across platforms.
Accomplishments that we're proud of
We’re proud of the final product, which successfully merges a wide range of technologies into a seamless, cloud-hosted pipeline.
What we learned
We learned a great deal about the streaming ecosystem, despite not being content creators ourselves. We also deepened our understanding of RTMP communication, video processing, and handling MP4 files in real time.
What's next for ClipDaddy
We aim to release a stable version in the near future and begin offering this service to streamers with large followings—those who stand to benefit most from automating and reclaiming their content distribution.
Log in or sign up for Devpost to join the conversation.