Postdoctoral Fellow: Bayesian ML for Biomedical Modeling

Commonwealth of VA Careers; Blacksburg (VA); On-site; USD 59,000 - 72,000; Full time

Job summary

The Virginia Tech National Security Institute (VTNSI) seeks applications for a Postdoctoral Associate to conduct research in computational modeling and uncertainty quantification as part of a multidisciplinary program.

The successful candidate will help develop surrogate models of computational systems via reduced-order and machine learning approaches and leverage Bayesian methods to estimate parameters, quantify uncertainty, and identify training needs to improve models.

Qualifications

• PhD in Applied Mathematics, Computational Science, Mechanical Engineering, Aerospace Engineering, Engineering Mechanics, or related field, awarded within last 4 years.
• US Citizenship due to project sensitivity.
• Experience with Bayesian inference and uncertainty quantification.
• Experience with computational modeling of physical or engineering systems.
• Programming experience in Python, MATLAB, Julia, C/C++, or similar.

Responsibilities

• Develop surrogate models of computational systems using reduced-order and machine learning methods.
• Apply Bayesian methods to estimate parameters and quantify uncertainty.
• Identify training points for model evaluation to improve accuracy.
• Publish peer‑reviewed work and prepare scientific presentations.
• Mentor students involved in related research and collaborate across institutions.

Skills

Bayesian inference

Parameter estimation

Uncertainty quantification

Computational modeling

Scientific programming

Education

PhD in Applied Mathematics or related field

Tools

Python

MATLAB

Julia

C/C++

Job description

The Virginia Tech National Security Institute (VTNSI) seeks applications for a Postdoctoral Associate to conduct research in computational modeling and uncertainty quantification as part of a multidisciplinary program.

The successful candidate will help develop surrogate models of computational systems via reduced-order and machine learning approaches and leverage Bayesian methods to estimate parameters, quantify uncertainty, and identify training needs to improve models.