-
-
OpenStream
Inspiration
The inspiration for OpenStream OTT came from exploring how fragmented and inaccessible global live TV content can be. While many free public streams exist across the world, they are often scattered, poorly organized, and difficult for users to explore in a clean and modern way. At the same time, I was interested in understanding how generative AI tools like Gemini could accelerate real-world application development beyond simple demos.
I wanted to build something practical and user-facing that combined a modern OTT-style experience with rapid AI-assisted development, showing how ideas can move from concept to deployment quickly.
What it does
OpenStream OTT is a web-based OTT-style application that allows users to discover and stream publicly available live TV channels from around the world. It provides a clean interface for browsing channels and playing live streams directly in the browser, without requiring accounts, subscriptions, or logins.
The application focuses on accessibility, simplicity, and global reach, turning raw IPTV streams into a usable and modern streaming experience.
How we built it
The project was built using a modern frontend stack with React, TypeScript, and Vite. The application is deployed on Vercel for fast, global access and automatic redeployments on code changes.
Gemini 3 was used extensively during development as an AI-assisted engineering tool. It helped with application architecture, component design, streaming logic, debugging, and deployment configuration. Using Gemini significantly reduced iteration time and allowed rapid experimentation with UI structure, folder organization, and best practices.
The final result is a production-ready frontend application deployed and accessible via a public URL.
Challenges we ran into
One of the main challenges was handling live stream playback reliably in the browser, as different streams behave differently depending on format and availability. Another challenge was setting up a clean Git and deployment workflow, ensuring the project structure was scalable and suitable for future expansion.
Accomplishments that we're proud of
- Successfully deployed a fully functional OTT-style web application
- Created a public, accessible demo without paywalls or logins
- Used Gemini 3 meaningfully to accelerate real-world application development
What we learned
This project reinforced the importance of clean architecture, honest AI usage, and iterative development. I learned how generative AI can act as a powerful development accelerator, helping with reasoning, debugging, and design decisions, even when AI is not directly part of runtime inference.
I also gained hands-on experience with modern frontend tooling, deployment pipelines, and best practices for presenting real-world projects.
What's next for OpenStream OTT
Future plans include adding optional Gemini-powered runtime features such as AI-generated channel descriptions, intelligent categorization, multilingual explanations, and smarter discovery tools. A lightweight backend could also be introduced to preprocess stream metadata and enable more advanced AI-driven features.
The project is designed to evolve from a streaming aggregator into an intelligent discovery platform powered by Gemini.
Log in or sign up for Devpost to join the conversation.