Notes on craft, design, and the engineering work behind it. Long-form posts, short essays, and project debriefs.
Floods are among the most damaging natural hazards, and they fall hardest on places with the fewest river gauges and the weakest warning systems. Machine learning is now forecasting floods …
Insurance is a business built on data and probability, which makes it a natural home for machine learning. AI is speeding up claims, sharpening underwriting and making micro-insurance viable for …
AI governance has moved from conference panels into product roadmaps. The EU AI Act, the NIST AI Risk Management Framework and a growing list of national policies now shape what …
Should you call a frontier API or host an open-weight model yourself? The honest answer depends on data sensitivity, latency, cost curves and your team's operational maturity. This guide breaks …
Not every AI model belongs in the cloud. TinyML puts inference on microcontrollers that sip milliwatts, enabling smart sensors that work offline, protect privacy and cost very little. For regions …
Africa is home to a vast share of the world's languages, yet most AI systems still serve only a handful of them well. Closing that gap requires more than bigger …
