Thought leadership, technical deep-dives, and field notes from the intersection of data science and real-world impact.
Kenya has 107,000 community health workers generating millions of data points every month — yet most of that data sits unanalysed. Here's what needs to change.
How we built a malaria prediction model that works with incomplete, delayed, and low-quality data from rural health facilities.
A frank look at the gap between what donors want to see and what program teams actually need to make decisions.
What we learned from trying to use remote sensing data to identify the highest-need farmers across three countries.
Technical lessons from deploying AI assistants for community health workers in East Africa where connectivity is unreliable.
How we operationalize ethical AI in high-stakes environments — and why most ethical AI frameworks fall short in the field.
How psychometric data, mobile behavior, and social capital signals can predict creditworthiness where financial records don't exist.