https://youtu.be/K4GFj_6iNbk
<aside>
🏆
Eventum is a production, local-first handball match tracking and analytics platform used by 200+ active recurring coaches across four countries. I am its Founder & Lead Software Engineer, responsible for the product and its complete technical lifecycle.
Open the Eventum web
</aside>
What I Build and Own
- Product architecture: Define the system boundaries, domain models, data flows, package structure, and long-term technical roadmap.
- Full-stack implementation: Develop the Flutter applications, shared Dart core, Firebase integration, cloud synchronization, analytics services, and supporting backend workflows.
- Offline-first data layer: Design Hive persistence, repository abstractions, migrations, user-scoped storage, and recovery behavior.
- Distributed collaboration: Implement shared live matches, multi-editor event streams, organization workspaces, role-based access, and synchronization under unstable network conditions.
- Analytics engine: Build team and player statistics, xG models, heatmaps, momentum calculations, tag insights, dashboards, and report generation.
- Quality and delivery: Maintain automated tests, Firestore security-rule coverage, CI/CD pipelines, release configuration, localization, and multi-platform builds.
- Product development: Translate direct coach feedback into features, UX improvements, pricing tiers, and technical priorities.
System Architecture
Eventum is a Dart pub workspace organized as a monorepo:
eventum_core — UI-independent domain models, Hive persistence, repositories, authentication, Firestore services, and reusable business logic.
eventum_live — the production Flutter application for live match capture, analysis, organizations, and reporting.
eventum_video — the video-analysis application and shared event-to-video workflows.
The application follows strict MVVM + Riverpod layering:
Views → ViewModels → Services → Repositories → Hive / Firestore
Views contain no business logic. ViewModels coordinate reactive state, services implement domain workflows, and repositories isolate persistence and cloud behavior.
Local-First Persistence and Synchronization
- Hive is authoritative: Every match action is persisted locally first, providing zero-latency interaction and full offline operation.