Winter School in AI and Predictive Agriculture

The Winter School in AI and Predictive Agriculture is a free event to share the collective learnings of experts from The University of Queensland in the use of AI and predictive analytics in Agriculture and Sustainability. The event will be held on 13-15 July 2026 at The University of Queensland St Lucia Campus, Brisbane and will feature seminars, case studies and hands-on workshops.

The Winter School aims to:

• Build capacity in applied AI for agriculture, especially amongst PhD students, EMCRs and industry participants;
• Provide participants with practical skills, not just knowledge;
• Demonstrate AI as an enabler to make standard agricultural technologies more accessible.

In person registrations for this event have closed, you can still register to attend online via this form >

Monday 13 July: AI Toolkit and Foundations

All day: Lecture, online or in person. Morning tea and lunch will be provided for in person attendees.
Attendees will learn the foundations of AI in an accessible way to support the applied case studies and hands‑on workshops in the following days. Topics will include:

• Intro to Neural Networks
• AI Models and Embeddings
• Transformers and LLMs
• Agentic AI
• AI tools deployment

Tuesday 14 July: Phenotyping

AM: Lecture, online or in person. Morning tea will be provided for in person attendees.
Attendees will explore how the AI Toolkit can be applied to advanced predictive phenotyping through expert seminars and real-world case studies. Topics will include:

• Sensors and point clouds
• Machine learning for phenotyping
• Foundation models for crops
• High throughput platforms and 3D digital twins
• UAV image data processing

PM: Hands-on workshop, in person only (spaces limited). Lunch will be provided.
Participants will apply the methods introduced on real-world problems.

Wednesday 15 July: Genomic prediction

AM: Lecture, online or in person. Morning tea will be provided for in person attendees.
Attendees will learn how AI methods can be leveraged for genomic prediction via expert seminars and applied case studies. Topics will include:

• Fundamentals of Genotype-to-Phenotype models
• Ensemble models
• Graph Neural Networks applied to genomic prediction
• Marker selection using biological priors and networks
• Agentic coding to build a genomic prediction pipeline

PM: Hands-on workshop, in person only (spaces limited). Lunch will be provided.
Participants will apply the methods introduced on real-world problems.

Spaces for hands-on workshops are strictly limited.

Scientific and Technical Committee

David Kainer

Senior Research Fellow, The University of Queensland, ARC Centre of Excellence for Plant Success

Genomic prediction and graphs.

Liqi Han

Research Fellow, Queensland Alliance for Agriculture and Food Innovation

Crop Digital Twins, High Performance Computing, Artificial Intelligence, High Throughput Phenotyping.

Brodie Lawson

Postdoctoral Researcher, Queensland University of Technology, ARC Centre of Excellence for Plant Success

Crop growth modelling across scales, including plant signalling, cell-growth and physiology. Gaussian process regression as a surrogate to true modelling to create fast models for crop simulators.

Christos Mitsanis

PhD Student, The University of Queensland, ARC Centre of Excellence for Plant Success

Agentic AI, mechanistic networks and CGMs.

Sophie (Qiaomin) Chen

Research Fellow, The University of Queensland

Multi-scale precision agriculture. Integrates crop modelling with machine learning. modelling of crop traits using simulation-generated data sets.

Sebastian Lopez-Marcano

Senior Manager Analytics, The University of Queensland

Data analytics, workflows.

Javier Fernandez

Research Fellow, The University of Queensland

Crop and agronomic modelling.

Katie O’Connor

Senior Plant Breeder, Queensland Department of Primary Industries

Plant Breeding, using genomic tools to improve selection methods.

Eric Dinglasan

Senior Research Fellow, Queensland Alliance for Agriculture and Food Innovation

Optimisation strategies for QTL stacking models. Adapts methods proven in animal breeding to crops.