As we look to agricultural systems to meet the ever increasing demand in food, fuels, and fibers over the coming decades, the challenges are immense. If viewed as a manufacturing process, farming is unique in its substantial temporal disconnect between providing inputs to, and realizing outputs from the system, and in its exposure to biophysical variability. Gaining a better understanding of the myriad forms of variability in the farming process, thereby reducing risk to producers and enabling ecologically sound land management, is the central challenge facing the agricultural sector today.
I’ve been fortunate to be able to view agricultural problems from multiple domains, and from the perspective of agriculturalists in both developed and developing economies. I’ve worked with consulting firms, technology providers, agribusinesses, research institutions and NGOs, in a variety of roles. Through this process, I’ve developed a passion for building models – economic, statistical, and otherwise – to better characterize farming systems and agricultural landscapes in order to enable more informed decision making processes.
Most recently, I’ve begun the long journey of acquiring data science and machine learning tools in order to apply them to agricultural problems. I’ve never been more excited or determined about anything professionally. The simultaneous growth in data availability, in the form of remotely sensed and instrument-derived agronomic data, along with the application of sophisticated machine learning algorithms, present an opportunity to dramatically improve agricultural productivity, improve livelihoods for resource-poor farmers via access to insurance, value chains, and financial services, and enable land management from an agroecological perspective. I’m eager to be a part of this transformation.