Geochemistry Postdoctoral Fellow, Lawrence Berkeley National Laboratory

Berkeley Lab’s  Earth and Environmental Sciences Area (EESA) has an opening for a Postdoctoral Fellow to join their Geochemistry Department!


You will work on a project that involves developing new data-based models and theories of reactive multicomponent Earth-subsurface systems.  In this role, you will work closely with computer scientists and molecular modelers to extract new knowledge/develop models of the complex geological systems using machine learning methods.  The research project will involve machine learning and statistical analysis of synthetic data sets generated using the state-of-the art molecular simulations.   

Multi-component aqueous solutions, and mineral/solution interfaces are common in energy-relevant Earth, atmospheric and technological systems, frequently under extremes of temperature and pressure. It is a major challenge to predict the chemical thermodynamics and chemical kinetics of such complex fluids.  Inadequate models for these systems limit our ability to predict and control processes in diverse energy and water systems.  Recent developments in data science are driving a paradigm shift in many areas of science and technology, ranging from image and speech recognition to drug design. 


What You Will Do:

Co-design, implement, and test machine learning models to extract new knowledge from the molecular simulation data (correlations, responses, relationships, data engineering, feature extraction).

Implement chemistry- and physics-based layers as constraints on machine learning models to develop interpretable ML approaches.

Perform fundamental analysis of the simulation data, collect synthetic database using molecular modeling approaches to the problems relevant to subsurface geoscience systems.

Present research results through preparation of articles and manuscripts for peer-reviewed journals.


What is Required:

PhD in Computer Science, Applied Mathematics, Data Science or relevant Physical Sciences discipline.

Demonstrated technical skills with relevant software including machine learning (tensorflow, scikit-learn, Keras, Caffe) or molecular modeling (NWChem, DL_POLY, Lammps, Gromacs, VASP, cp2k). 

Demonstrated ability to lead, complete and publish scientific research projects.

Excellent oral and written communication skills.


What We Desire:

Experience applying machine learning or molecular modeling algorithms.


The posting shall remain open until the position is filled.

Deadline
31 December 2019
Back to top