1,000 beta users and a recommendation engine that kept them scrolling
2025
Me taking home an award for DLICIO
Demo Reel
DLICIO project demo
DLICIO is a short-form food platform with ML-powered dish recommendations. I built the recommendation engine and backend. We hit 1,000+ beta users and 4,000+ waitlist signups. After CraveMatch shipped, average session duration jumped 50% (3.2 → 4.8 minutes). Probably the metric I'm most proud of.
Stack
DINOv2——Self-supervised ViT used as a visual feature extractor. Strong out-of-the-box dish similarity without labeled training data
Spark ALS——Collaborative filtering over user interaction history, combined with DINOv2 embeddings to generate personalized recommendation scores
React Native——Cross-platform short-form feed with restaurant profiles and order flow, matching TikTok-style UX on both iOS and Android
Docker——Each service (ML inference, recommendation engine, order management, frontend) containerized independently for per-service auto-scaling
AWS——Independent auto-scaling groups per container, API Gateway for routing and rate limiting, CloudWatch for latency monitoring
