Summer Data Intern: Kiwifruit Conditioning Predictor

Zespri Group Limited; Mount Maunganui; On-site; NZD 67,000 - 100,000; Full time

Benefits offered by this job

Free onsite parking

Full project exposure

Job summary

Zespri invites a curious and analytical Summer Intern to join our Conditioning Team in Mount Maunganui. You’ll help develop a data-driven tool that predicts kiwifruit softening by integrating time, temperature and ethylene exposure, combining data science with postharvest research.

Working alongside our team, you’ll analyse historical data, review literature, build predictive models and deliver a practical decision-support tool with clear recommendations for conditioning decisions across the

Qualifications

• Currently pursuing or recently completed a science/technology/horticulture or business degree at NZQA level 6-10
• Programming capability in Python or R and interest in postharvest science or data-driven decision tools

Responsibilities

• Review scientific literature and historical conditioning data to identify key drivers of fruit firmness.
• Analyse datasets relating to time, temperature, ethylene exposure and fruit softening.
• Develop a unified conditioning metric combining multiple variables.
• Build and validate models to predict firmness.
• Test and refine models with historical and current-season data.
• Deliver a practical decision-support tool for operational use.
• Present findings, recommendations and future opportunities for development.

Skills

Python or R

Analytical thinking

Data analysis

Education

Undergraduate or postgraduate degree related to science/tech/horticulture or business

Tools

Python

R

Job description

Zespri invites a curious and analytical Summer Intern to join our Conditioning Team in Mount Maunganui. You’ll help develop a data-driven tool that predicts kiwifruit softening by integrating time, temperature and ethylene exposure, combining data science with postharvest research.

Working alongside our team, you’ll analyse historical data, review literature, build predictive models and deliver a practical decision-support tool with clear recommendations for conditioning decisions across the