Research Scientist

Key responsibilities

• Design, fit and interpret mixed-effects, hierarchical and longitudinal models for structured agricultural data (repeated measures; nested farm/county/region effects; genetic, environmental and management variance components).
• Build predictive and forecasting models (time-series, panel regression, gradient boosting and ensemble methods) for production, demand, price and market-intelligence questions at a national and sub-national level.
• Develop population and value-chain projection models (herd dynamics, yield response, supply–demand balances) that feed policy, investment and sector-planning analyses.
• Lead the analytical design of data-analytics projects, producing insights that drive evidence-based policymaking and private-sector decisions.
• Collaborate with government, research partners and industry stakeholders to frame analytical questions and resolve data challenges in agriculture.
• Contribute statistical models and outputs to digital tools and dashboards developed with the data engineering team.
• Prepare technical reports, peer-reviewed publications and visualizations that communicate findings to diverse audiences.
• Ensure adherence to data governance standards, ethical AI principles and best practices in reproducible data management.
• Work closely with the pillar lead to refine methodologies, improve model performance and scale analytics solutions.

Candidate requirements

• Minimum Requirements.
• Master’s or PhD in Statistical/Quantitative Genetics, Crop or Animal Breeding, Agricultural Economics, Biostatistics, Statistics or a closely related quantitative field. Candidates with a Data Science or Computer Science background will be considered only with demonstrated applied experience in agricultural or biological research and data analytics.
• Demonstrated expertise in mixed models (e. g. lme4/nlme, ASReml, SAS PROC MIXED or equivalent), generalised linear models, and predictive/forecasting methods.
• Strong expertise in machine learning and predictive analytics, with sound judgement on when statistical inference versus algorithmic prediction is appropriate.
• 3+ Years’ experience in applied statistical analysis or quantitative research.
• 3+ Years’ experience in agricultural data collection, management and analysis (livestock, crop, farm-survey, market or trade data).
• Evidence of applied output: peer-reviewed publications, technical reports or models that have informed real policy, investment or operational decisions.
• Proficiency in R, STATA, SAS and/or Python for statistical modelling; familiarity with dashboard tools (e. g. Power BI, Tableau) is an advantage.
• Demonstrated ability to work with large, messy, multi-source datasets and derive actionable insights.
• Desirable (added advantage)
• Breeding-value estimation, genomic prediction or variance-component estimation in livestock or crops.
• Agricultural economics modelling: partial-equilibrium or CGE models, supply-response or demand-system estimation.
• Bayesian methods (e. g. Stan, INLA, brms) and spatial or geospatial statistics.
• Experience with remote-sensing or GIS data in agricultural applications.
• Experience working with government or development-partner data systems in Africa.

Required skills