Inspiration
Short-form videos are everywhere, but understanding why different videos perform differently can be time-consuming. Creators often switch between YouTube Shorts, Instagram Reels, analytics pages, and notes to research content. We built ClipScope AI to bring this scattered information into one workspace and make content research more organized, accessible, and evidence-driven.
What It Does
ClipScope AI is an AI-powered short-form video research and analytics platform. Users can organize supported YouTube Shorts and Instagram Reels, review available video metadata, compare videos side by side, analyze transcripts, explore AI-generated observations, and ask questions through a research chat. It also provides workspace management, analytics dashboards, background processing, and exportable reports.
How We Built It
The application uses Next.js, React, TypeScript, and Tailwind CSS for its web interface. Supabase provides authentication, PostgreSQL storage, and access control. YouTube Data API and a configured Instagram data provider supply available video metadata. Groq powers AI-generated observations and research chat, while Inngest supports background processing, retries, and recovery. The architecture separates retrieved platform data from AI-generated interpretations to make the results easier to evaluate.
Challenges We Faced
One of the main challenges was handling incomplete and inconsistent data across platforms. Video metadata and transcripts are not always available, particularly for Instagram Reels. We designed the system to distinguish unavailable information from actual zero values and to avoid inventing missing transcripts or metrics. Other challenges included coordinating asynchronous jobs, managing workspace-level access, handling retries, and keeping AI responses grounded in the available evidence.
What We Learned
Building ClipScope AI helped us understand how to combine external APIs, database security, background workflows, and language models into a practical application. We also learned that useful AI analysis requires more than generating convincing text: data availability, transparent limitations, and traceable evidence matter just as much.
What's Next
Our next goal is to extend ClipScope AI with an Amazon developer technology integration, enabling a more conversational way to explore content research and performance insights. We aim to make the platform more useful for creators and content teams while keeping recommendations grounded in available data rather than promising viral outcomes.
Built With
- groq
- inngest
- next.js
- postgresql
- react
- supabase
- tailwind-css
- typescript
- vercel
- youtube-data-api
