Your data skills directed at the problems that actually need solving. Climate, health, agriculture, and financial inclusion in Jamaica and the Caribbean.
Data for Good is not about learning data science in a vacuum. It is about pointing your skills at real problems: the flooding in Kingston, the maternal mortality rates in rural parishes, the small farmers who lose income because they cannot predict weather patterns.
Participants work in small teams alongside community organizations, government data departments, and NGOs. You learn data science by doing data science for people who need it. Every project has a real-world partner and a real-world impact target.
This program runs continuously. You can join an active project team at any intake point, depending on which challenge areas need support right now.
Pick the area that pulls you. Each has ongoing projects looking for contributors with varying skill levels.
Jamaica is on the front line of climate change. Rising sea levels, intensifying hurricanes, shifting rainfall. We use satellite data, ocean temperature records, and atmospheric models to build early warning tools and resilience dashboards for communities and local government.
From predicting diabetes risk in underserved parishes to mapping mental health service gaps, data can guide where resources go and who gets help first. We partner with health clinics and the Ministry of Health on specific projects.
Jamaica imports too much of its food. We work with the Ministry of Agriculture and small farmer cooperatives to build crop planning tools, soil health trackers, and yield prediction models using local data and remote sensing.
Hundreds of thousands of Jamaicans are unbanked or underserved by financial systems. We analyze credit access patterns, remittance flows, and mobile money usage to help fintech organizations and credit unions serve people better.
A community-facing dashboard mapping flood risk by parish using topographic, rainfall, and land-use data. Now used by two parish councils for infrastructure planning decisions.
A lightweight ML model identifying communities with high diabetes risk based on diet, activity, and socioeconomic indicators. Deployed at three rural health clinics.
A seasonal yield prediction model for yam farmers in St. Elizabeth using weather forecasts, historical yields, and soil data. Pilots showed 28% reduction in planning waste.
No waiting for the next cohort. Tell us which challenge area interests you and your current skill level. We will match you with an active team.