About Me

I am a Data Scientist and incoming PhD Researcher who builds end-to-end AI systems to translate complex data into measurable, real-world outcomes.

Commencing in April 2026, I will be undertaking a PhD in Artificial Intelligence at The University of Queensland, supported by a fully funded RTP scholarship and a SAAFE CRC top-up. My research will focus on developing Cross-Sectoral Digital Twins for Antimicrobial Resistance (AMR) risk assessment.

I hold a Master's in Analytics (Health) from Massey University and am concurrently completing a Master's in Computer Science (AI) at Monash University. My work sits comfortably at the intersection of data science and software engineering — I enjoy not just researching models, but building robust infrastructure that actually ships.

Robert Spencer

Areas of Expertise

Bridging the gap between research and production-ready AI systems

  • Machine Learning & AI

    Designing and deploying proprietary ML pipelines, from research to production. Experience with Transformer architectures, transfer learning, and time-series forecasting.

  • Digital Twins

    Developing cross-sectoral Digital Twin frameworks for complex system modelling, with focus on AMR risk assessment and building energy consumption forecasting.

  • Data Engineering

    Building robust data pipelines and software infrastructure. Integrating multiple databases, encoding regulatory logic, and creating end-to-end ML workflows.

Top Skills

  • Synthetic Data Generation
  • Antimicrobial Resistance
  • Digital Twins Modelling
  • Python & ML Engineering
  • Time Series Forecasting

Experience & Education

Professional Experience

Lead Data Scientist

Radix Nutrition | Jan 2025 - Oct 2025
Brisbane, Australia

  • Led end-to-end design and deployment of AI-driven nutrition optimisation systems, cutting formulation iteration time by 90%+ (from weeks to days)
  • Led data-driven substantiation for 150+ health claims, enabling key marketing positions including 'World's Healthiest'
  • Co-developed AI-optimised breakfast and smoothie range, reducing formulation costs by ≥25%
  • Built essential in-house tools — dashboards, GUIs, and data pipelines — adopted by R&D and Operations teams

Data Analyst → Data Scientist

Radix Nutrition | Apr 2024 - Jan 2025
Technical hire #1 — established entire data infrastructure

  • First technical employee responsible for establishing company's entire data and software infrastructure
  • Tackled data analytics, software engineering, web development, and nutritional science

Research Assistant — AI

Massey University | Sep 2023 - Oct 2024
Palmerston North, New Zealand

  • Conducted novel research on transfer-learning for building energy consumption forecasting
  • Achieved 15.9% reduction in Mean Absolute Error for 24-hour forecasts using multi-source dataset approach
  • Analysed 16 diverse building datasets from Building Data Genome Project 2
  • Demonstrated superior performance of PatchTST architecture vs vanilla Transformer and Informer models
  • Resulted in first-author Q1 publication in Energy and Buildings journal

Education

PhD in Artificial Intelligence

The University of Queensland | Commencing April 2026
Fully funded RTP scholarship + SAAFE CRC top-up
Research: Cross-Sectoral Digital Twins for AMR Risk Assessment

Master of Computer Science (AI)

Monash University | 2024 - 2027 (Concurrent)
Specialisation in Artificial Intelligence

Master of Analytics (Health)

Massey University | 2023 - 2024
Specialisation in Health Analytics

Bachelor of Health, Sport & Human Performance

The University of Waikato | 2020 - 2022
Double Major: Human Performance Science & Community Health

Certifications

  • Google Data Analytics Certificate
  • SAS Academic Specialisation — Massey University, Health Analytics

Publications

Transfer learning on transformers for building energy consumption forecasting — A comparative study

Spencer, R., Ranathunga, S., Boulic, M., van Heerden, H., & Susnjak, T. (2025)
Energy and Buildings (Q1 Journal)

This study presents a comprehensive comparative analysis of transfer learning applications using Transformer architectures for building energy consumption forecasting. We achieved a 15.9% reduction in Mean Absolute Error for 24-hour forecasts using an innovative multi-source dataset approach across 16 diverse buildings.

Contact

I love collaborating with researchers and practitioners in artificial intelligence, analytics, applied health, and sustainable innovation.
If you are interested in the intersection of AI and complex biological or environmental systems, I'm always keen to connect.