
SoulSync
An AI-powered Spotify playlist generator that creates personalized playlists based on user input using LLMs and real-time music data.
Timeline
2-3 Days
Role
Full Stack Developer
Team
Solo
Status
CompletedTechnology Stack
Key Challenges
- Prompt Engineering for Song Generation
- Parsing LLM Output Reliably
- Spotify API Integration
- Accurate Song Matching
- CORS & Deployment Handling
Key Learnings
- LLM Integration in Real-world Apps
- API Orchestration
- User-driven Personalization
- Frontend-Backend Integration
- Data Cleaning from AI Outputs
Overview
SoulSync is an AI-powered playlist generator that creates personalized Spotify playlists based on user preferences such as mood, artists, or context. Instead of relying on predefined rules, it uses an LLM to dynamically generate song recommendations and maps them to real Spotify tracks.
The system combines AI-generated recommendations + Spotify API execution, resulting in fully playable playlists created in real-time.
Key Features
Core Functionalities
- AI Playlist Generation: Generate playlists based on mood, text, or preferences
- LLM Integration: Uses Gemini to suggest relevant songs
- Spotify Integration: Automatically creates playlists in user’s account
- Real-time Processing: Instant playlist creation
- Multi-source Intelligence: Combines LLM + web search for better recommendations
- Clean UI: React-based frontend for smooth interaction
How It Works
- User Input → Mood / description / artist preferences
- AI Processing → Gemini generates song recommendations
- Web Search Agent → Enhances recommendations with external context
- Parsing Layer → Extracts song names & artists
- Spotify Matching → Finds closest tracks via Spotify API
- Playlist Creation → Creates playlist in user account
- Output → Returns playlist link
System Flow (Simplified)
- LLM → Song Suggestions
- Regex Parsing → Structured Song List
- Spotify Search → Track IDs
- Playlist API → Final Playlist
API Endpoint
POST /generate_playlist