About the Program

This internship is hosted within WRI’s Data Lab, the organization’s core data innovation and product delivery unit. The Data Lab supports work across Institute — including the Food, Land, and Water (FLW) Program and Land & Carbon Lab—by providing technical infrastructure, geospatial analytics, and applied data science. The Land and Forest initiatives and the Supply Chains Hub within FLW aim to conserve and restore forests and other ecosystems critical to people, nature, and the climate. This internship will contribute to those efforts by supporting agricultural and forest-monitoring analytics through data labeling and geospatial workflows.

This is a hybrid position which requires 8 days per month in the office. You can be based in our office in Washington DC. Existing work authorization is required at the time of application submission as WRI is unable to sponsor any visa work sponsorship for this position. To be eligible for this position, you must be a resident of DC, Maryland or Virginia at the start of employment.

What you will do:

You will be responsible for integrating equity into your work.

  • Assist in the development and application of agricultural field boundary datasets to support local use cases in Latin America and Africa
  • Support annotation and digital sketching tasks to generate training datasets for deep learning models for these use cases
  • Gain experience working within a global environmental NGO on geospatial data innovation
  • Contribute to models at the field boundary level, such as yield estimation, natural conversion, and/or supply chain risk assessment
  • Collaborate with a distributed, international team of GIS analysts, researchers, and data scientists
  • Work across WRI’s country offices to apply spatial insights to sustainability, food security, and land use challenges

    You will be supported by data scientists and land use specialists in the Data Lab, Food Land and Water Program, and Land and Carbon Lab. You will report to the Agricultural Data Scientist in the Data Lab.

    Internship Learning Outcomes:

    • Learn to annotate and create high-quality polygon datasets for deep learning model training.
    • Expand knowledge of agricultural practices and how to characterize them through analysis.

      You will be embedded within WRI's Data Lab, the core technical unit providing strategy, infrastructure, and innovation for data-intensive projects. You will specifically support efforts to expand and apply agricultural field boundary data and analytics to country office initiatives in Colombia, Brazil, Mexico, and Kenya. This work bridges GIS, agriculture, and sustainability, supporting high-impact projects ranging from identifying deforestation-free production zones to improving smallholder crop yield mapping.

      Internship Duties:

      Labeling and Annotation (60%):

      • Use digital tools to annotate field boundaries and agricultural features in satellite imagery (Sentinel-2, Planet)
      • Follow structured standards and quality control practices for training and validation data generation
      • Prioritize labeling tasks across use cases in Colombia, Mexico, Kenya, and Brazil
      • Collaborate with experts from WRI’s Toolkit for Traceability (TkT) and partner research groups

        GIS Analysis and Content Development (30%):

        • Assist with GIS processing of field boundaries and land cover datasets using tools like Geospatial Python (geopandas, GDAL, rasterio), ArcPro, and Google Earth Engine for supply chains, cadastral dataset analysis, land use analysis, and other use cases
        • Conduct spatial analysis to support country-office needs (e.g., crop identification, agricultural frontier mapping, bioeconomy analysis)
        • Contribute to simple data visualizations and cartographic products for internal and external use

          Documentation (10%):

          • Contribute to documentation of workflows and help build reproducible annotation pipelines
          • Engage with WRI country office staff to understand analytical needs and translate them into labeling or GIS tasks

            What you will need:

Recommended for you