ML
& DEVOPSENGINEER
FULL STACK
CLOUD INFRA

What I Do

I build inference
pipelines, ship
cross-platform apps,
and deploy custom
models to the edge.
From cloud to silicon,
end to end.

Projects

A selection of projects I have built, from full-stack apps to developer tooling and ML pipelines.

01

Ragadex

Cross-platform mobile application shipped to both the App Store and Google Play. Transforms live music festivals into a collectible card game — passive artist card acquisition via precision GPS geofencing with zero user interaction. Resilient offline-first architecture with deterministic background sync and conflict resolution.

Native MobileOffline SyncGeofencingPostGIS
02

Heatmap Core Inference Engine

End-to-end sports intelligence platform coupling custom-trained detection and tracking models with a production analytics dashboard. Processes raw multi-camera court footage through a proprietary inference pipeline to produce granular player heatmaps, kinematic trajectories, and derived performance metrics. Containerized microservice architecture with automated schema migrations.

Custom ModelsCV PipelineSpatial AnalyticsMicroservices
03

Embedded ML Courtside System

Purpose-built edge inference appliance running proprietary fine-tuned YOLO object detection models on a Raspberry Pi 5 with dedicated AI HAT accelerator. Dual-camera RTSP ingestion with real-time stream processing, automated temporal segmentation, and orchestrated cloud persistence to object storage.

Edge InferenceCustom YOLONPU AcceleratorRTSP Streams
04

News Intelligence Agent

Autonomous intelligence agent that systematically ingests news corpora, performs entity extraction and sentiment analysis via foundation model inference, and synthesizes structured executive briefs. Architected as a fully serverless event-driven pipeline with decoupled processing stages and durable message orchestration.

Agentic AILLM InferenceServerlessEvent-Driven