AI Movie Recommender

Inspiration

The overwhelming amount of content available on streaming platforms can make it difficult for users to find movies they'll truly enjoy. I was inspired to create a solution that leverages AI to provide personalized movie recommendations, making it easier for users to discover films tailored to their unique tastes and preferences.

What it does

The AI Movie Recommender analyzes users' past movie preferences and interactions to suggest films they are likely to enjoy. It uses advanced sentiment analysis and Retrieval-Augmented Generation (RAG) with the latest LLaMA AI model to provide insightful movie summaries and recommendations. Users receive a weekly email with a curated list of movies tailored to their tastes, keeping them engaged and excited about their next movie night.

How I built it

I built the AI Movie Recommender using Next.js for the frontend and Node.js for the backend, integrated with the LLaMA 3.1 model via API for AI-driven recommendations. I utilized the TMDb API to fetch movie data dynamically and employed sentiment analysis to fine-tune recommendations based on user reviews. The project is deployed on Vercel, ensuring a seamless and scalable deployment.

Challenges I ran into

One of the biggest challenges was integrating the LLaMA AI model for real-time recommendation generation, particularly ensuring the AI's responses were contextually accurate and relevant to the user's movie preferences. Additionally, fine-tuning the sentiment analysis to reflect nuanced user feedback posed another challenge, as did optimizing the overall performance for a smooth user experience.

Accomplishments that I'm proud of

Despite the time constraints and working solo, I successfully integrated advanced AI capabilities into a user-friendly movie recommendation system. My ability to dynamically fetch and analyze movie data, combined with delivering personalized recommendations via email, marks a significant achievement. I'm particularly proud of the seamless user experience and the scalability of the platform.

What I learned

This project taught me a great deal about integrating advanced AI models like LLaMA into a real-world application. I gained valuable experience in managing API interactions, handling complex data processing, and deploying AI-driven solutions at scale. I also honed my skills in ensuring a seamless frontend-backend integration.

What's next for AI Movie Recommender

Due to time limitations in my life, I was only able to complete the initial version of this project in one day. I plan to continue polishing and expanding the features, including enhancing the recommendation capabilities to cover TV shows and other forms of media. I aim to refine the sentiment analysis further to capture more subtle user preferences and explore additional AI models to improve recommendation accuracy. My ultimate goal is to create a comprehensive, AI-powered media recommendation platform that caters to all entertainment needs.

Built With

• api
• javascript
• llama
• mui
• next.js
• rag
• react
• tailwind