For a YouTube live streamer, there was a clear problem: viewers often caught amazing moments on stream but had no easy way to save or share them without scrubbing back through hours of footage. The creator wanted something lightweight that empowered the audience to create clips during the livestream itself, helping turn viewers into active contributors rather than passive watchers. The goal was a fast MVP that could plug directly into existing YouTube livestreams without forcing anyone to learn complex editing tools.

In the modern creator economy, short-form video (TikTok, YouTube Shorts, Instagram Reels) is the primary driver of organic growth. However, extracting high-retention clips from long-form content (podcasts, streams, interviews) is a massive bottleneck.

Creators and agencies spend hours manually scrubbing through footage to find engaging hooks, reframing horizontal video to vertical (9:16), and manually animating captions. This manual workflow is expensive, unscalable, and prone to human error, drastically limiting a creator's output volume.

SnipeMe is an automated, AI-driven video clipping tool designed to turn hours of long-form video into viral, ready-to-publish short-form clips in minutes.

By leveraging advanced speech-to-text models and AI sentiment analysis, SnipeMe automatically identifies the most engaging moments in a video, crops the frame to keep the speaker centered, and overlays highly engaging, kinetic typography.

Ingest — Users paste a YouTube link or upload a raw video file.

Transcribe & Analyze — The AI generates a highly accurate transcript and analyzes the context to pinpoint "viral moments" (high energy, clear hooks, complete thoughts).

Smart Crop — The system uses facial recognition and active-speaker detection to reframe traditional 16:9 landscape footage into perfect 9:16 vertical clips.

Caption & Style — Kinetic, dynamic captions (similar to the "Alex Hormozi style") are automatically rendered onto the video with highlighted keywords and emojis.

Export — Users receive a dashboard of 10+ viral clips, complete with AI-generated titles and descriptions, ready for one-click download.

90% Reduction in Editing Time: What used to take an editor 3 hours now takes an AI less than 5 minutes.

Maximized Content ROI: Allows podcasters, agencies, and streamers to extract 10x more value from a single piece of long-form content.

Lowered Barrier to Entry: Empowers creators with zero video editing experience to produce professional, algorithm-friendly content instantly.

Building a heavy video-processing application requires a delicate balance between server costs and processing speed. By optimizing FFmpeg rendering pipelines and utilizing asynchronous API calls for the AI analysis, SnipeMe successfully delivers high-fidelity video processing natively in the cloud, resulting in a seamless and magical user experience.

Using React.js and Supabase, a full‑stack web app was built at snipeme.xyz that lets live viewers capture specific moments from an ongoing stream and turn them into short clips. The frontend focuses on a clean, responsive experience so anyone in chat can quickly mark and trim a highlight while the stream is still live. Supabase handles authentication, clip metadata, and storage, keeping the flow snappy and reliable while making it easy to manage clips over time.

The result is an audience-driven clipping tool that helps streamers get more shareable content with almost no extra effort on their side. Viewers can instantly save hype plays, funny moments, or key reactions, which can then be reused across YouTube Shorts and other platforms to grow the channel. This project showcases the ability to design around community interaction, ship a focused MVP quickly, and use modern tools like React and Supabase to support real-time creator workflows.