The Mission

Prometheus is building computer vision infrastructure for the people who manage Africa's land — conservancies, game reserves, farmers, and cooperatives. These stewards oversee landscapes spanning hundreds of thousands of hectares, yet still rely on manual aerial surveys, ground patrols, and guesswork to understand what's happening on the ground. Our mission is to turn drone and satellite imagery into continuous, reliable intelligence about the living systems these landscapes contain. We are starting where the pain is sharpest and our team has the deepest traction: aerial wildlife detection for conservancies. Rangers and conservation scientists will be able to automatically detect, classify, and count animals across species and terrains — spotting population anomalies, tracking migration shifts, and detecting poaching pressure earlier. The same underlying vision stack extends naturally into the adjacent challenges African land stewards face every day: crop health monitoring, rangeland degradation, soil condition, and livestock management. One platform, many land uses, one continent underserved by existing tools.

The Challenge

Monitoring land at scale remains a deeply unsolved problem across Africa — whether you are counting elephants in a 500,000-hectare conservancy or tracking crop stress on a smallholder farm. Manual aerial wildlife counts are costly, dangerous, and happen once or twice a year at best, leaving long blind spots during which poaching, disease, drought, and migration anomalies go undetected. Farmers face the mirror image of the same problem: no affordable way to know what is happening across their fields until yields are already lost. Existing computer vision tools fail these users. Wildlife detection models are typically trained on narrow datasets from single reserves or species and do not generalize to the diverse terrains, lighting, and species mixes found across African ecosystems. Agricultural remote-sensing platforms are built for European or North American conditions, priced for commercial operations, and rarely adapted to African crops or smallholder plot sizes. The organizations that need these tools most — conservancies operating on donor funding, and farmers operating on thin margins — are precisely the ones current solutions ignore.

The solution

Prometheus is an aerial intelligence platform for African land stewards. Users upload drone footage or connect satellite imagery feeds, and our system returns annotated imagery, dashboards, and anomaly alerts tailored to their use case. We are launching with wildlife detection for conservancies as our wedge. Under the hood, we fine-tune state-of-the-art object detection architectures (YOLOv11) on the open WAID aerial wildlife dataset and adapt them to each reserve's specific terrain and species mix through partnerships like the Malilangwe Trust in Zimbabwe. This transfer-learning approach means we can deploy reliably in a new reserve within weeks, not months, and improve continuously as more field data is collected. The same computer vision and geospatial ML stack extends directly into our next verticals: crop health and yield monitoring using Sentinel-2 satellite imagery, soil and rangeland condition assessment using multispectral indices, and livestock monitoring for commercial ranches and cooperatives. Every reserve we onboard strengthens the dataset and models that power the agricultural side, and every farm we eventually serve deepens our geospatial pipeline. One platform, shared infrastructure, compounding moat — built specifically for African conditions from day one.