Carl McMillan

Profile

---

Software engineer and AI researcher with over ten years of experience in software development and DevOps, and a recently submitted PhD in Artificial Intelligence. My research applies deep learning and computer vision to mussel farm monitoring, and has produced four first-author peer-reviewed publications. I enjoy turning AI research into practical tools that are useful in everyday tasks, and I am driven by solving problems in simple and efficient ways.

Skills

---

Languages

Python, Bash, Java, SQL

Deep Learning

PyTorch, CNNs (ResNet, DenseNet), Vision Transformers (ViT), YOLO object detection

Computer Vision & Data Science

OpenCV, scikit-learn, NumPy, pandas, Matplotlib

Cloud, HPC & DevOps

AWS, HPC clusters (Slurm, GPU), Terraform, Docker, Kubernetes, GitLab CI, Jenkins, Git, Linux

Work Experience

---

Doctoral Researcher at Victoria University of Wellington

 

Jul 2023 - Sep 2026

• Developed deep learning models (CNNs, Vision Transformers) for predicting the buoyancy of mussel farm floats from video, including BuoyancyNet, published in the Journal of the Royal Society of New Zealand.
• Built an end-to-end pipeline from raw mussel farm video to float buoyancy alerts, combining object detection and buoyancy prediction models.
• Assessed the feasibility of deploying the alerting system on a solar-powered monitoring buoy.
• Ran large-scale training and evaluation experiments on a Slurm-managed GPU cluster.

Research Assistant at Victoria University of Wellington

 

Apr 2023 - Jun 2023

• Annotated a mussel float detection and buoyancy dataset used by multiple researchers for the Centre for Data Science and Artificial Intelligence aquaculture research group at VUW.
• Trained an object detection model on this dataset and produced a video demonstration showcasing the Centre's capabilities.

DevOps Engineer at Fronde

 

Sep 2020 - Sep 2022

• Built AWS infrastructure for integrated microservice environments using Terraform.
• Built GitLab CI/CD pipelines for a monorepo microservices project.
• Set up monitoring and alerting using AWS CloudWatch, Prometheus, Grafana and Slack integrations.

Technical Consultant (Development/DevOps) at Westpac

 

Jun 2017 - Sep 2020

• Built Java microservices in Kubernetes, using Kafka for distributed messaging, to integrate proprietary bank card processing software running on older platforms.
• Implemented encrypted storage of financial transaction data in databases and Kafka topics for PCI DSS compliance.
• Automated server builds and installation of proprietary bank card processing software using Chef and Jenkins.
• Set up monitoring and alerting for these systems using Elastic Stack, Grafana and PagerDuty.
• Provided on-call production support for these systems.

Software Developer at Datacom

 

Jun 2011 - Jun 2017

• Developed web-based data visualisation software for tracking the availability of Chorus resources.
• Developed integration components for the Chorus fibre ordering system as part of the development team.
• Developed AMS integrations for remote operations on smart meters.
• Provided on-call production support for the team's existing applications.
• Improved the team's development processes through automation and ChatOps.

Software Developer at Wialus

 

Nov 2009 - Feb 2010

• Summer of Code internship.
• Created C++ programs for ETL processes for mobile phone signal data.
• Created a C++ program for calculating elevations along any great circle path.

Biosecurity Officer at GWRC

 

Jul 2006 - Nov 2007

• Controlled pest animals in the Wellington region and helped the public with pest animal problems.
• Set up poison and trapping programs for volunteers to take over.
• Carried out mapping with GPS devices and ArcGIS software.
• Maintained the database and generated reports for a large conservation project in the Wainui water catchment.
• Conducted trials for possum control methods.

Education

---

Victoria University of Wellington

 

Submitted Sep 2026

• PhD in Artificial Intelligence
• Thesis: Deep Learning for Building a Buoyancy Alerting System in Aotearoa New Zealand Mussel Farms

Victoria University of Wellington

 

Completed Oct 2023

• Postgraduate Diploma in Science in Artificial Intelligence

Victoria University of Wellington

 

Completed Dec 2011

• Conjoint BSc/BCA in Computer Science/Mathematics/Money & Finance

Victoria University of Wellington

 

Completed May 2004

• BSc in Ecology & Biodiversity/Marine Biology

Publications

---

• Carl McMillan, Junhong Zhao, Bing Xue, Ross Vennell, and Mengjie Zhang. Local Context-Aware Buoyancy Prediction for Mussel Farm Floats. PRICAI 2025: Trends in Artificial Intelligence, Wellington, New Zealand, November 2025, LNCS vol. 16452, pp. 459-474. DOI: 10.1007/978-981-95-7072-0_31
• Carl McMillan, Junhong Zhao, Bing Xue, Ross Vennell, and Mengjie Zhang. BuoyancyNet: A Deep Learning Approach for Assessing Float Buoyancy in Mussel Aquaculture. Journal of the Royal Society of New Zealand, vol. 55, no. 6, 2025, pp. 2013-2041. DOI: 10.1080/03036758.2025.2488415
• Carl McMillan, Junhong Zhao, Bing Xue, Ross Vennell, and Mengjie Zhang. Deep Learning-Based Buoyancy Prediction for Intelligent Mussel Farm Monitoring. 2024 39th International Conference on Image and Vision Computing New Zealand (IVCNZ), Christchurch, New Zealand, December 2024, pp. 1-6. DOI: 10.1109/IVCNZ64857.2024.10794479
• Carl McMillan, Junhong Zhao, Bing Xue, Ross Vennell, and Mengjie Zhang. Improving Buoy Detection with Deep Transfer Learning for Mussel Farm Automation. 2023 38th International Conference on Image and Vision Computing New Zealand (IVCNZ), Palmerston North, New Zealand, November 2023, pp. 1-6. DOI: 10.1109/IVCNZ61134.2023.10343657