Fynda was Kordspace’s internal concept for AI-assisted music and event matchmaking. It explored how musical preferences could inform discovery across both artists and live experiences, rather than treating those as separate recommendation problems. The concept centered on helping listeners find relevant options beyond a simple genre match. Python provided the technical foundation for exploring the relationship between preference data and recommendation logic. The project belongs to Kordspace’s internal AI prototyping work, rather than a commissioned client delivery.
The challenge
Music and event discovery involves subjective preferences that do not translate neatly into fixed categories. A useful matchmaking concept needs to connect musical affinity with practical event relevance while working with potentially incomplete or inconsistent information. The central design challenge was to make those connections understandable without presenting a recommendation as a guarantee of personal enjoyment.
Our approach
The prototype scope centered on a Python-based recommendation workflow connecting listener preferences, music attributes, and event information. A suitable data model would distinguish musical affinity from contextual factors such as location and event format, allowing each to influence a match independently. Candidate recommendations could then be ranked using explicit preference signals before introducing more complex learning techniques. Lightweight explanations would help communicate why an artist or event appeared relevant. This approach kept the exploration focused on matchmaking logic and data requirements rather than assuming a production application or autonomous agent architecture.
The outcome
Fynda represents an internal exploration of AI-assisted discovery across music and events. Its value lies in framing a product direction around connected recommendations, with a clear distinction between preference matching and practical event suitability. The project is best understood as a concept and prototype, not a validated commercial service or a demonstrated production deployment.
Technical highlights
- Python-centered AI prototyping
- Cross-domain recommendation design connecting music and live events
- Separation of preference signals from practical event constraints
- Explainability-oriented matchmaking concept
Place in the journey
2017–2019 Software + CloudFynda illustrates Kordspace’s exploration of data-driven product concepts alongside its broader application engineering work. It connects AI experimentation with the practical product question of how to make discovery more personally relevant.