The Quadracci Sustainable Engineering Lab (QSEL) works on solutions that could advance well-being and economic growth through both access to energy, water and infrastructure as well as creating and enabling income generating opportunities in emerging economies. In recent years, we have increasingly been asked to help identify opportunities and binding constraints at granular scale using national-scale data of populations, land-use, economic activity and infrastructure to support the ability of private sector and governments to leverage opportunity and address constraints. The incumbent will be contributing to ‘Using Data to Catalyze Energy Investments’, a Columbia World Project.
The primary job responsibilities of this position are to:
1) implement deep learning models for identifying cropland, seasonal changes in vegetation, distinguish between horticulture and cereal crops and possible use of irrigation. Inform on value of variety of label data sets and time series radar imagery as well as daytime high-resolution imagery.
2) develop associated code base for multi-sensor image processing and analysis,
3) test, validate and scale-up predictions,
4) document methodology and write publications,
5) work closely with collaborators and practitioners.
- Ph.D. in computer science, data science, environmental science, engineering, landscape ecology or a closely related field;
- Extensive experience in machine learning for image analysis;
- Strong programming skills;
- Excellent written communication skills demonstrated by prior publications; and
- A track record that demonstrates the ability to work well with interdisciplinary research teams.
- Experience with processing multi-spectral satellite imagery; computer vision.
- Strong programming skills in Python, Pytorch, KERAS, TensorFlow; and
- Experience with cloud-computing tools for geospatial computing (e.g. Google Earth Engine) will be considered advantageous.