Post-doctoral Researcher - Adaptive Façade Intelligence and Control Systems

Qualifications: PhD in Artificial Intelligence, Machine Learning, Computer Science, or a related discipline

Term: One-year, full time, fixed-term position

Start date: 15 October 2026

Location: Wellington or remote (International applicants are encouraged to apply)

Remuneration: NZ$84,000 per annum

About the Project

The Biomimetic Buildings project is a three-year research programme funded through the Ministry of Business, Innovation and Employment (MBIE) Smart Ideas programme. More information on the project can be found here.

We are seeking a post-doctoral researcher to support the development of intelligent, adaptive façade systems for climate resilient and energy-efficient buildings. The role will focus on designing, developing, implementing, and validating machine-learning models and control algorithms for the automated, real-time operation of dynamic building shading systems.

The research will use existing building-performance simulation data and multi-objective optimisation results to train and validate machine-learning models and control policies. These will support computationally efficient, real-time façade operation that maintains indoor environmental quality (e.g. occupant comfort) while minimising operational energy use.

A central component of the role will be developing control strategies that can respond appropriately when an occupant specifies a desire for a different environmental state. Following an occupant’s expressed desire for more/less light, warmer/colder conditions, reduced glare etc., the control system will use real-time sensor measurements, system feedback, and an independently developed control policy to identify the next-best feasible action or sequence of actions to change the façade configuration. This will guide the façade back towards energy-efficient operation and the required indoor environmental quality targets without requiring a complete online optimisation process at every control step.

The successful candidate will have expertise in machine learning (and more specifically deep learning), sequential decision-making, automation, and closed-loop control, with the ability to apply these methods to responsive façade systems. The candidate will work closely with a Mechanical/mechatronics Engineer (who is being recruited simultaneously to be part of the Biomimetic Buildings project team) responsible for developing and testing the physical façade system.

Key Responsibilities

• Develop and validate machine-learning models (e.g. artificial neural networks) and control algorithms using simulation-generated data and multi-objective optimisation results to assess façade performance in terms of thermal conditions, airflow, daylight, glare, and energy use under different façade configurations and operating conditions.

• Develop and evaluate control strategies for selecting appropriate façade configurations under changing environmental and operational conditions.

• Design control methods that respond to occupant input and guide the façade back towards optimised operation.

• Integrate real-time sensor data and system feedback into closed-loop control and automation processes.

• Assess the accuracy, robustness, computational efficiency, and real-time suitability of the developed models and control methods.

• Contribute to technical documentation, research publications, presentations, and project reporting.

Essential Skills and Experience

• Strong knowledge of machine-learning methods, including artificial neural networks, deep learning, and reinforcement learning.

• Experience with developing machine-learning models for dynamic systems and applying them within optimisation or control frameworks.

• Experience developing supervised machine-learning models. 

• Understanding of sequential decision-making, feedback control, closed-loop system operation, and real-time control.

• Experience developing machine-learning models that approximate computationally expensive simulations or optimisation processes.

• Proficiency in Python and relevant scientific-computing and machine-learning libraries, such as PyTorch, or TensorFlow.

• Ability to evaluate model performance, identify overfitting, quantify uncertainty, and assess the generalisation of models to previously unseen operating conditions.

• Ability to write clear technical documentation and communicate research methods and results effectively.

• Ability to work collaboratively within a multidisciplinary research and development team.

• Familiarity with GitHub, or equivalent version-control and collaborative software-development tools. 

Why Join Us?

• Contribute to the development of sustainable and high-performance building technologies.

• Work within a collaborative, multidisciplinary, and inclusive research environment.

• Access professional-development opportunities through research, publication, conference participation, and international collaboration.

• Benefit from flexible working arrangements, noting that for this project you will be working remotely.

• Use advanced computational tools to support your work.

• Join a supportive team that values innovation, intellectual curiosity, humility,  and continuous learning

Application Process

Applications must be emailed to Negin Imani at negin@bodekerscientific.com by midnight on 24 September 2026.

Please provide:

• A covering letter.

• A curriculum vitae that includes:

  • your experience and strengths in relation to the essential skills described above;

  • links to relevant GitHub repositories or examples of your code, where available; and

  • the names and contact details of at least two referees.

Short-listed applicants will be invited to an interview, either in person or via Zoom.