A machine learning specialist in horticulture focuses on applying data-driven algorithms and models to optimise agricultural practices. They work with large datasets, including climate conditions, soil health, crop yields, and other environmental factors, to develop predictive models and improve decision-making processes. Their role is essential in advancing precision agriculture, automating tasks, and enhancing crop management for better efficiency and sustainability.

These professionals leverage advanced machine learning techniques to analyse trends, predict outcomes, and automate processes such as pest detection, irrigation management, and harvest forecasting. By integrating machine learning into horticulture, they help businesses increase productivity, reduce resource waste, and adapt to changing environmental conditions.

Core tasks

• Develop and implement machine learning models for agricultural optimization.
• Analyse large datasets to identify trends and insights for improved crop management.
• Create predictive models to forecast crop yields, pest outbreaks, and irrigation needs.
• Automate agricultural tasks, such as monitoring, pest detection, and irrigation scheduling.
• Collaborate with horticulture professionals to integrate machine learning solutions into practices.

Career pathways

Due to the technical nature of this field, most occupations require formal study, such as:

• Bachelor of Science in Computer Science or Data Science
• Master’s in Machine Learning or Artificial Intelligence

Q&A with Annie Wang, Machine Learning Team Lead, Bitwise Agronomy

Q: What is it you actually do?

A: I manage a team of machine learning engineers and data scientists at Bitwise Agronomy, and we focus on developing deep learning models for fruit counting. We also develop AI-driven yield forecasting and design practical computer vision solutions to address the needs of growers.

Q: What would you say is the most rewarding part of working in this industry?

A: I would say the most rewarding part is to see the direct impact of how AI could address those real challenges in this industry.  Whether it’s using AI for improving yield forecasting or automating fruit counting, it is just really satisfying to see the solutions that we create being applied straight into production.

Q: Can you share a memorable project or experience that you’ve had while working in your field?

A: One of the most memorable projects that I worked on was developing a fruit-counting algorithm using only video data. Counting fruit in videos is just horribly challenging because the fruits often look very similar and are really difficult to distinguish. 

But we have come up with a very smart way of doing that by just using detections on video frames, without using any extra data such as GPS, which I know some other companies rely on. We were actually able to achieve around 90% accuracy, which was a great achievement for our team.

Q: What advice would you give to someone like that considering a career in the industry?

A: Always stay curious and open-minded. I think this field is just growing so fast, and there’s so much room for innovation.