Researcher in Machine Learning and MLOPS (2 Year Full Time Fixed-term Contract) at University of the Witwatersrand

• The position will be hosted within the School of Statistics and Actuarial Science at the University of the Witwatersrand, which hosts a growing research programme in AI for climate risk modelling and climate intelligence, supported in part by the Bezos Earth Fund. This programme focuses on developing machine learning methods for weather forecasting, climate risk modelling, and climate-impact analytics relevant to African institutions.
• The successful candidate will work at the intersection of machine learning engineering, atmospheric science, and large-scale geospatial data systems, supporting the development and deployment of machine learning models for weather and climate prediction. The role will focus on building scalable ML pipelines, managing large climate datasets, enabling reproducible experiments, and supporting model training and evaluation on high-performance computing and cloud infrastructure.
• The Researcher will work closely with researchers and students within the FineCast project at the University of the Witwatersrand, collaborating with partners across AfriClimate AI, African National Meteorological and Hydrological Services (NMHSs), and international research institutions.
• This position is a 2-year full-time fixed-term appointment

Brief Description        
Requirements

• A Master’s degree or PhD in Computer Science, Machine Learning, Data Science, Software Engineering, Applied Mathematics, or a related field.
• Strong programming skills in Python and experience with modern machine learning frameworks such as PyTorch, TensorFlow, or JAX.
• Experience building and maintaining machine learning pipelines and data workflows.
• Experience working with large-scale scientific or geospatial datasets.
• Experience with high-performance computing environments, GPU clusters, or cloud platforms (e.g., GCP, AWS, or similar).
• Experience using version control systems (e.g., Git) and collaborative software development practices.
• Ability to work effectively in interdisciplinary research teams.
• Strong problem-solving skills and attention to reproducibility and reliability in scientific computing workflows.  
• Experience in one or more of the following areas will be advantageous:
• MLOps and machine learning infrastructure
• Distributed training of deep learning models
• Geospatial data processing and climate datasets
• Containerization technologies such as Docker and Kubernetes
• Weather or climate modelling systems
• Forecast verification or climate data analytics

Key Responsibilities

• Develop and maintain machine learning infrastructure and pipelines supporting the FineCast research programme.
• Build scalable workflows for training, evaluating, and deploying AI-based weather and climate forecasting including generation, compression and  archival of  hindcasts from frontier models.
• Manage and process large meteorological and climate datasets, including satellite data, reanalysis products, and observational station data.
• Support the training and fine-tuning of global AI weather prediction models using regional datasets.
• Implement systems for experiment tracking, reproducibility, and model versioning in machine learning research.
• Develop tools and infrastructure supporting forecast verification and model benchmarking.
• Support the integration of weather forecasting outputs with climate risk modelling and analytics workflows developed within the School’s research programme.
• Contribute to the development of open-source software and research tools produced by the project.
• Work closely with researchers and postgraduate students to translate research ideas into scalable and reliable machine learning systems

Deadline:18th October,2026