GrooveTrack
A dance learning experience that won Most Creative Approach at HackSprint 2025.
GrooveTrack made learning a physical skill visible: a dance journey with progress, recommendations, and an award-winning creative premise.
The situation.
HackSprint ’25 asked builders to create the website or app they wished they had when starting a hobby. The team focused on a common dance-learning gap: people can find moves, but may lack a structured journey and a clear way to see progress.
GrooveTrack proposed personalised dance journeys, guides, progress tracking, and exercise or stretch recommendations. The project story explicitly records that an initially ambitious AI recommendation idea was simplified to a rule-based system within the hackathon timeline.
What I did.
I built GrooveTrack with Leewon Min, Arina Babikyan, and Anan Osman. We shared the product, engineering, and presentation work, and won Most Creative Approach as a team.
My contribution sat inside that collaborative build, helping turn the dance-learning idea into a working journey that we could demonstrate within the hackathon.
How the work unfolded.
Defined a structured learning loop
Connected move guides, progress tracking, and dynamic suggestions based on skill level and goals.
Balanced ambition with time
Replaced a stretched AI recommendation engine with a rule-based approach that could be integrated and demonstrated reliably.
Integrated the prototype stack
The submission documents a React/ShadCN/Tailwind frontend, Django REST backend, and Firebase for data and media.
What happened.
GrooveTrack won Most Creative Approach at HackSprint ’25, recognising the team’s creative approach to making dance progress visible.
Replacing the original AI recommendation idea with a rule-based system helped us finish a coherent demo within the hackathon. The final build connected dance guides, progress tracking, and exercise recommendations without depending on a rushed feature we could not make reliable in time.
What I learned.
Creative feedback can be concrete
Progress in a physical skill can be made legible through movement goals, practice guidance, and a visible journey, not only through productivity metrics.
Simplification can protect the experience
Choosing a rule-based recommender let the team finish a working loop instead of stretching the build around an unfinished AI system.
Awards credit the submission, not a single maker
The Most Creative Approach result belongs to the named team and should remain attributed that way.
Related links.
Repositories, demos, event pages, posts, and other places connected to this project.