Remote Spatial Data Scientist

1h1 hour ago

CIFOR-ICRAF

Remote · Full-time · $85,000 – $120,000

About this role

This remote role at CIFOR-ICRAF focuses on spatial data science, leveraging machine learning and remote sensing for ecosystem health assessment. You will design machine learning pipelines for geospatial analysis, including feature engineering, model selection, and validation. Develop deep learning models like CNNs, RNNs, LSTMs, and Transformers for image tasks and time series forecasting.

Day-to-day involves processing optical data from Sentinel-2 and Landsat, SAR from Sentinel-1, with data fusion for ML workflows. Implement time series analysis for vegetation, precipitation, and land dynamics, including trend detection and anomaly identification. Build scalable, reproducible workflows and contribute to MLOps practices.

Supervise junior spatial data scientists and developers while leading capacity development seminars on ML, AI, and spatial data science within CIFOR-ICRAF. Conduct workshops for partners emphasizing ML-driven spatial analysis. Collaborate with the SHARED stakeholder engagement team on AI outputs for projects like the Great Green Wall.

Engage stakeholders in decision support tools and contribute to micro-dashboards for the Global Resilience Impact Tracker. Support projects with analytical outputs and decision-maker engagement. Lead or contribute to scientific papers and proposal development.

Requirements

PhD or MSc degree in spatial data science, geoinformatics, computer science, or a related quantitative field with demonstrated expertise in machine learning and AI applications
Proven experience developing and deploying machine learning models for geospatial applications
Strong proficiency in deep learning frameworks (TensorFlow, PyTorch, Keras) and familiarity with architectures such as CNNs, RNNs, LSTMs, and Transformers
Advanced programming skills in Python and/or R; familiarity with Julia is a plus
Experience with cloud computing platforms (GEE, AWS, GCP) and big data processing tools for geospatial analysis
Knowledge of remote sensing data processing and analysis, including optical and SAR platforms
Excellent interpersonal skills
Excellent written and spoken English; knowledge of French a plus

Responsibilities

Design and implement machine learning pipelines for geospatial analysis, including feature engineering, model selection, hyperparameter tuning, and validation
Develop and deploy deep learning models (CNNs, RNNs, LSTMs, Transformers) for image classification, segmentation, object detection, and time series forecasting
Process and analyze optical data (Sentinel-2, Landsat 8/9) and SAR data (Sentinel-1), including data fusion and feature extraction for ML workflows
Implement time series analysis and forecasting models for vegetation, precipitation, and land surface dynamics
Develop scalable, reproducible spatial data processing workflows and contribute to MLOps practices
Supervise a team of junior spatial data scientists and developers
Lead internal capacity development seminars on machine learning, AI applications, and spatial data science
Work closely with the stakeholder engagement team to provide AI-driven analytical outputs for project delivery