Experience
TuMi Mundowith Alfa Meerwaarde, NWO, University of Leiden & Sydney
Svelte, JavaScript, Directus, D3.js, Pico.js face tracking library
Helped build an interactive tool for TuMi Mundo, tracking babies’ attention and encouraging language development through audio & stimulating visuals, while ensuring secure and anonymous data handling.
Click here to view the projectOverview
TuMi Mundo is an interactive research platform designed to study and encourage early language development in infants. The tool tracks babies' visual attention and engagement in real time while providing stimulating audio-visual cues.
Project snapshot
- My role: Front-end developer
- Team: Researchers from Alfa Meerwaarde and the University of Leiden
- Type: Project during the Minor Web Design and Development
- Stack: Svelte, JavaScript, Directus, D3.js, Pico.js face tracking library
The challenge
The main challenge was to build a reliable, real-time interactive system for infants, while ensuring all data remained secure, anonymized, and consistent. The interface needed to be engaging but non-distracting, and the frontend had to work seamlessly with the research backend.
The solution
We implemented the frontend using Svelte for reactive performance and D3.js to visualize real-time tracking data. The Pico.js library was integrated to monitor facial attention, and all interactions were logged securely via Directus.
My contributions
- Built reusable interactive components for attention tracking and visual/audio stimuli
- Integrated Pico.js for facial attention detection and real-time data capture
- Collaborated with researchers to translate experimental protocols into frontend interactions
- Worked on responsive layouts to ensure consistency across devices
- Ensured secure, anonymized data handling while maintaining performance
Challenges & learnings
This project taught me how to build highly interactive, research-focused web tools that need to balance performance, usability, and data privacy. I gained experience translating research requirements into frontend functionality and learned best practices for building reliable, real-time systems in a sensitive context.
Outcome
TuMi Mundo successfully enabled researchers to collect and visualize infants’ attention data interactively. The platform remained performant and secure, while providing an engaging interface for study participants.
