University of Cambridge · Mastercard Foundation Scholar
I am passionate about improving and protecting human health and livelihoods by combining deep learning with infectious-disease modelling and translating evidence into policy that reaches those who need it most.
About
Gerald Munetsi is a researcher working at the intersection of deep learning and infectious-disease modelling. He is a Mastercard Foundation Scholar and has completed an MPhil in Population Health Sciences at the MRC Biostatistics Unit, University of Cambridge, having previously earned a first degree in Applied Mathematics from the University of Zimbabwe.
He is currently an Assistant Research Fellow at the Centre for Epidemiology Modelling and Analysis (CEMA), where he contributes to the following projects: Dynamic Transmission Modelling for Schistosomiasis and Soil-Transmitted Helminths for the WHO Expanded Project for Elimination of Neglected Tropical Diseases (NTDs), and Mathematical Modelling to Establish the Transmission Dynamics and Burden of Visceral Leishmaniasis in Kenya and Ethiopia to Inform Targeted Control and Elimination Strategies (ENDVL Afrika).He is also currently implementing a Policy Fellowship awarded by the Africa Pandemic Sciences Collaborative, focusing on AI-enhanced One Health surveillance.
His research interests centre on:
Notable Work
Making computationally expensive agent-based epidemic simulators fast enough to answer real-time policy questions.
Health Data Research UK · United Kingdom.
Deep-learning approaches to Bayesian inference over pathogen genomes · United Kingdom.
Applied optimisation for smallholder agriculture · Zimbabwe.
Working with policymakers from the Institute of National Planning, Egypt to generate evidence across the ministerial conditions and governance frameworks required to prototype an African-owned, AI-enhanced One Health surveillance system — enabling district officers to act on zoonotic spillover signals in real time.
University of Cambridge
United States
Research & Code
A sample of open work — from epidemic emulators to genomic classifiers. Full source is on GitHub.
Surrogate models that emulate expensive stochastic epidemic ABMs, so inference and scenario analysis run in seconds rather than hours.
View on GitHub →Simulation-based / neural inference for phylodynamic and genetic-epidemiology models, learning posteriors over pathogen transmission from genomic data.
View on GitHub →Multi-class classification of cancer type from high-dimensional genomic expression data using machine-learning pipelines.
View on GitHub →Comparative risk-assessment modelling of dietary policy scenarios and their projected population-health impact.
View on GitHub →Linear/optimisation models to solve the land-allocation and transport problem for small-scale farming in Zimbabwe.
View on GitHub →Notebooks, experiments and research code across epidemiology, ML and statistics.
@GeraldMunetsiGallery
Cambridge, conferences, community and the road here.
CLEAR-Africa Consortium · Oxford, 2026
Contact
Open to research collaborations, speaking, policy work and partnerships across AI, epidemiology and global health. The fastest way to reach me is email or LinkedIn.
germunetsi15@gmail.com