Projects
I build things to solve real problems. Some are polished. Some are messy. All of them taught me something.
Weather & Meteorology
SpiceWeather A modern weather platform for Grenada, designed to serve the public, aviation, and internal operations from a single coherent system.
Ensemble Visualisation Tools Python-based tools for exploring and comparing ensemble forecasts (ECMWF, GFS, NZLAM), including heatmaps and station-based summaries. Built to help forecasters see uncertainty instead of hiding it.
Climate & Observation Databases Systems built with Python and PostgreSQL to validate, store, and transform observational data — supporting climatology, aviation reporting, and long-term analysis.
Early Warning & CAP Work (pilot / concept) Explorations into Common Alerting Protocol workflows and impact-based warning systems aligned with international standards.
Software Engineering
Weather & Aviation APIs REST APIs built with FastAPI to deliver forecasts, observations, alerts, and derived products to web and mobile clients.
Web Applications React and Next.js applications focused on clarity, speed, and usability — because scientific data deserves good interfaces too.
Automation & Infrastructure Dockerised services, CI/CD pipelines, and cloud deployments designed to be boring, reliable, and easy to maintain.
Machine Learning (Emerging)
Pattern Discovery in Weather Data Applied clustering techniques to historical datasets to explore structure not obvious through traditional analysis.
AI-for-Weather Notes & Prototypes Early experiments, reading notes, and conceptual work around AI-based forecasting, nowcasting, and ensemble post-processing.
Each project has a story: what problem I was solving, what worked, what didn’t, and what I’d do differently next time.