Dark

DARK is an AI-powered Movie Intelligence Platform where Nyx reasons over user preferences, delivers explainable recommendations, and enables collaborative movie discovery through natural language.

Comment

Inspiration

Traditional movie recommendation platforms rely on static algorithms that often produce repetitive suggestions and provide little explanation for why content is recommended. Users also face fragmented experiences across multiple streaming services, making movie discovery and group decision-making time-consuming. We wanted to build more than another recommendation system—we wanted to create an intelligent AI agent capable of understanding user intent, reasoning over personalized viewing behavior, and taking meaningful actions within the platform. This vision led to DARK, an AI-powered Movie Intelligence Platform driven by Nyx, an agentic AI assistant designed to make movie discovery intelligent, explainable, and collaborative.

What it does

DARK is an AI-powered Movie Intelligence Platform that combines personalized recommendations, explainable AI, behavioral analytics, and collaborative features into a single experience.

At the heart of the platform is Nyx, an AI agent that understands natural language, reasons over user preferences, and dynamically interacts with the platform. Instead of functioning as a generic chatbot, Nyx can:

1.Understand user intent. 2.Answer platform-specific questions. 3.Recommend movies based on personalized viewing behavior. 4.Explain recommendations using Movie DNA and Taste Persona. 5.Guide users through platform features. 6.Help manage watchlists and collaborative movie planning. 7.Extend the same AI experience into Slack for team collaboration.

The platform also includes Movie DNA analytics, Taste Persona visualization, watchlists, streaming provider availability, trailers, search, and collaborative co-viewing.

How we built it

DARK was built as a full-stack web application using a modular architecture. Frontend- HTML5 CSS3 JavaScript Backend- Node.js Express.js Database- MongoDB Mongoose AI- Google Gemini API Nyx AI Operating System Intent Detection Context Builder- AI Gateway Structured Tool Routing Integrations- TMDB API,Slack API

Nyx follows an agentic workflow: User Query → Intent Detection → Context Building → AI Reasoning (Gemini) → Decision Making → Structured Response → Platform Action

Security features include JWT authentication, rate limiting, request validation, caching, secure API management, and Slack request verification.

Challenges we ran into

The biggest challenge was designing Nyx as an actual AI agent rather than a simple chatbot. We needed to build an orchestration layer capable of understanding user intent, selecting the correct platform capability, generating contextual prompts, minimizing AI latency, and maintaining secure responses without exposing internal system information. Other challenges included:

  1. Integrating multiple external APIs.
  2. Optimizing Gemini usage and response latency.
  3. Designing explainable recommendation systems.
  4. Building collaborative Slack integration.
  5. Preventing unnecessary AI requests through intelligent routing and caching.
  6. Balancing performance, scalability, and maintainability. ## Accomplishments that we're proud of
  7. Built a production-ready AI-powered Movie Intelligence Platform.
  8. Developed Nyx as an agentic AI operating system instead of a traditional chatbot.
  9. Integrated Google Gemini into a secure reasoning pipeline.
  10. Created explainable recommendation systems through Movie DNA and Taste Persona.
  11. Successfully extended Nyx into Slack for collaborative AI interactions.
  12. Built a responsive full-stack application with modern UI/UX. Implemented secure authentication, caching, rate limiting, and scalable architecture. Combined AI reasoning, analytics, collaboration, and personalization into one unified platform. ## What we learned Developing DARK significantly improved our understanding of building production-ready AI systems. We learned:
  13. How to design AI agents that reason and perform platform-specific actions.
  14. How to integrate Large Language Models into real-world applications responsibly.
  15. The importance of modular architecture for scalability.
  16. Techniques for reducing AI latency through context optimization and caching.
  17. How to secure AI applications using authentication, rate limiting, and request validation. 6.Best practices for integrating external APIs while maintaining reliability and performance.
  18. How to build explainable AI experiences that users can understand and trust. ## What's next for Dark Future plans include:
  19. Multi-agent collaboration within Nyx.
  20. Voice-enabled AI interactions.
  21. Cross-platform integrations beyond Slack.
  22. Advanced group recommendation and collaborative decision-making.
  23. Richer behavioral analytics and adaptive learning.
  24. Personalized notifications and proactive recommendations.

Built With

Share this project:

Updates

Submission history