Senior Geospatial Data Scientist

About Meridia Meridia is an agtech company built around the first mile: the plots, farmers and aggregation points at the origin of every agriculture supply chain. Our first mile intelligence combines what field teams see on the ground and what satellites see from above, in the context of local law, foreign regulation and international frameworks. It rests on over a decade of field experience, our data lake covering 20%+ of the world's agricultural land, a network of origin experts across the global South and a stellar team of 70+ agriculture, data and tech experts. Meridia Verify®, our verification solution, checks supply chain field data against regulatory and voluntary sustainability frameworks in minutes and guides teams from risk detection to resolution. One platform covers EUDR, Deforestation- and Conversion-Free (DCF) commitments and Scope 3 land-use emissions at plot level, and benchmarks a supply chain against sector baselines. Meridia Illuminate®, our sourcing intelligence solution, leverages the same first-mile data foundation to support secure sourcing decisions and resilient supply chains for the operators who shape global food and agriculture trade. See https://www.meridia.land for more information on our work. Introduction This document outlines the scope and objectives for hiring a consultant to support Meridia's first-mile intelligence work: the data science that connects commodity supply chains to the production areas they source from. This includes production area assessment, supply shed modelling and traceability, drawing on a variety of data sources and combining geospatial, statistical and ML approaches. Understanding where commodities come from, without needing every farm boundary, helps buyers make credible, evidence-based sourcing risk assessments, while keeping smallholders included in supply chains. It supports deforestation- and conversion-free (DCF), EUDR and Scope 3-related risk assessment. Objectives Develop accurate, high-quality models and datasets for production area assessment and sourcing area (supply shed) analysis across multiple commodities and origin countries Leverage Meridia's in-situ farm plot datasets to train and validate models Adapt existing machine learning models to local agricultural contexts to improve prediction accuracy Create scalable and reproducible methodologies that work across different geographies and commodities Deliverables Methodology and codebase Organised folder with all source files Full documentation including operating workflow, methodology, sources (including retrieval and licensing details) Requirements Must have Advanced degree (MSc or PhD) in Remote Sensing, Geospatial Science, Computer Science, or related field 10+ years of practical experience in applying statistical and earth observation techniques to agricultural applications Proficiency in programming languages commonly used in geospatial analysis and machine learning (Python, R, Google Earth Engine) Knowledge of agricultural systems and supply chains in tropical and subtropical regions Experience with modelling approaches relevant to supply chain analysis, such as optimisation, spatial allocation, entity matching or network analysis across heterogeneous datasets Strong geospatial data engineering skills: spatial databases (e.g. PostGIS), GDAL/geopandas or equivalent, and handling of large raster and vector datasets Experience building reproducible, version-controlled analytical pipelines (Git; containerisation such as Docker is a plus) Proven ability to document methodologies rigorously, including data provenance, assumptions, limitations and licensing Ability to work independently in a remote setting, with strong communication skills and full professional proficiency in English Nice to have Familiarity with agri-commodity sectors (e.g. palm oil, soy, rubber, cocoa, coffee) and their data landscape, including production statistics, mill or processor lists, and sourcing structures with producers and intermediaries Familiarity with (agricultural) supply chain transparency and traceability approaches, logistics, procurement and certification Working knowledge of relevant frameworks (EUDR, DCF/NDPE, certification schemes) and how production area and sourcing area information supports risk assessment under them Experience validating models against ground-truth or in-situ field data, and quantifying and communicating uncertainty Experience working with production statistics and yield data in data-scarce environments Track record of peer-reviewed publications or public technical documentation in a related domain Prior experience in Southeast Asia and/or Latin America Benefits This is an independent consultancy assignment rather than an employment role. Here is what you can expect from working with us: Initial 3-month engagement, with possible extension depending on progress and mutual fit 2-3 days per week Fully remote Software subscriptions provided by Meridia where applicable Monthly invoicing against agreed milestones and deliverables Work on a data lake covering 20%+ of the world's agricultural land, alongside a team of 70+ agriculture, data and tech experts Direct impact on EUDR, DCF and Scope 3 compliance work How To Apply If this opportunity matches your experience, please apply on Workable: https://apply.workable.com/j/739CC757DB Your application should include: CV highlighting relevant experience in geospatial data science and machine learning for agricultural applications A brief proposal (max 3 pages) outlining your approach Examples of previous work Your day rate and availability