Postdoctoral Researcher - Data Science and Machine Learning
The Residential Buildings Research Group is seeking a Postdoctoral Researcher with experience in machine learning applications to data science and controls for the built environment. The ideal candidate will leverage their scientific mindset and a creative “get it done” mentality to address important data-driven problems related to residential buildings, internet-of-things (IoT), building-grid integration, analysis of large and diverse datasets, and dynamic interactions of complex systems.
Job duties and responsibilities may include:
- Contribute to the development of methods and processes for assessing minimally-sufficient data/sensors for operating and control objectives in buildings
- Develop approaches for IoT enhancement of efficient buildings, especially system identification, statistical analysis and deep-learning methods
- Develop and apply time series data analysis methods to sub-hourly building energy meter and sensor data
- Design and validate aggregated benefits of distributed hierarchical control
- Further develop & collaborate to demonstrate probabilistic optimization of buildings and aggregation of statistically-defined grid services
- Mine large datasets to derive new learnings at the boundaries of building science, such as patterns of devices’ usage by occupants, methods for sizing home batteries, building stock trends, and other emerging issues.
- Developing novel optimization methods for complex, nonlinear and dynamic systems
- Writing papers for peer-reviewed science journals.
- Strong background in mathematics, statistics, machine learning, optimization or related fields is required.
- Demonstrated excellence in programming languages such as Python, R, SQL, Matlab or similar is required.
- Extensive experience with core Python libraries such as NumPy, SciPy, Pandas, Matplotlib and Scikit-Learn is required
- Experience with time series analysis methods (e.g., regression, classification, forecasting) is a plus.
- Working knowledge of data visualization tools and version control software is a major plus.
- Experience in big data and/or deep learning is a major plus (such as TensorFlow, Movidius, CNTK, etc.)
- Effective communication skills are required
- Proven track record of high-quality publications in peer-reviewed journals and conferences is required.
- Domain exposure to buildings or smart cities is a major plus.
- Domain exposure to utility power systems (the grid), electric vehicles, solar energy, battery storage, IoT or other NREL technologies is a plus
- Experience developing and using distributed cloud computing, especially secure multi-party computation, is a plus.
- Experience monitoring, mining, synthesizing from and analyzing multiple sources of digital and/or event-based data, such as network traffic and IoT data, is a plus.
Required Education, Experience, and Skills
Must be a recent PhD graduate within the last three years.