[CASE 01] // 2026
PITCHIQ
AI PRODUCT / FOOTBALL ANALYSIS PLATFORM
An analyst-grade football platform — upload a match, run an async video pipeline, and surface evidence-first tactical insights, timelines, and reports for coaches and scouts.
Role
Product design, full-stack architecture, video analytics pipeline
Services
WEB / STRATEGY
[CHALLENGE]
Football analysis tools often blur raw tracking data with AI interpretation, inventing stats or hiding uncertainty. Coaches and analysts need a workstation that separates what was detected from what it means — and scales from demo review to uploaded match video without breaking trust.
[APPROACH]
[01]
Built a dark analyst UI in Next.js with role-based access for coach, analyst, and player workflows.
[02]
Designed a modular async pipeline from preprocess through pitch calibration, detection, tracking, events, and tactical interpretation.
[03]
Stored raw tracking separately from interpreted insights and reports so evidence stays inspectable.
[04]
Shipped a demo match path for pitch, timeline, heatmaps, and AI report review without waiting on computer vision.
[05]
Abstracted external football data behind a provider interface with mock defaults for local development.
[OUTCOME]
A structured MVP for match upload, processing status, manual correction, and evidence-first reporting — ready to extend toward production video analytics without collapsing detections into black-box summaries.