KUL08 - AlgaWatch : AI-Powered Algal Bloom Prediction System

What if a fish farmer could see a deadly algal bloom coming days before it arrived, instead of waking up to a dead harvest? Today, that's not how it works.  Every year, harmful algal blooms sweep through coastal waters and devastate fish farms, suffocating cages, contaminating shellfish, and wiping out a season's income with almost no warning. The frustrating part? The data to see these blooms coming already exists — satellites watch the ocean from above, environmental sensors track conditions in the water, and ocean models map how blooms move. So why do people still get blindsided? Because today's warning systems lean on a single source of data, were built for very different waters, and stop at producing a forecast and leaving coastal communities to make blind, high-stakes decisions. AlgaWatch changes that. We fuse environmental data with satellite observations into one predictive model that forecasts and catches blooms before they reach the farm. Then we close the gap everyone else ignores: a decision-support layer that turns raw ocean data into clear, timely guidance people on the coast can actually act on. Going further, we're building an agentic system that reasons through the data to support decisions and adapt as conditions shift. The result is a leap from blind to forewarned which is an ocean intelligence once reserved for industrial producers, put in the hands of the communities who need it most.

A functional predictive system that leverage the uses of open marine data from different sources, combine with an agentic system that help in decision supports and help user to navigate easily on the system. It contains an interactive map dashboard that visualize the risk for a target coastal area and issues early-warning alerts. The core deliverables is the predictive model with the dashboard that turns its output into a clear decision .

AI Engineer, Machine Learning, Full Stack Developer, Electronics & Electrical Engineering, Cloud Infrastructure