Summary
The University of Pittsburgh is seeking a highly motivated Postdoctoral Researcher to join a three-year research project focused on applying artificial intelligence and machine learning (AI/ML) to subsurface energy challenges. The project, funded through the U.S. Department of Energy’s Science-informed Machine Learning to Accelerate Real Time subsurface decision making (SMART) LDRD Prime initiative, aims to develop a laboratory-calibrated AI/ML tool for estimating subsurface permeability from commonly collected geophysical well logs.
Postdoctoral Position in Machine Learning for Subsurface Multiscale Structure and Characterization at the University of Pittsburgh
The researcher will develop and validate advanced deep-learning models using unique datasets from national laboratories and collaborate with researchers across geophysics, geology, engineering, and computer science.
Summary Table
| Category | Details |
|---|---|
| Position | Postdoctoral Researcher |
| Department/Area | Geology and Environmental Sciences |
| Institution | University of Pittsburgh |
| Location | Pittsburgh, Pennsylvania, USA |
| Appointment | Three-year postdoctoral appointment |
| Employment Type | Full-time regular |
| Research Focus | Machine Learning, Subsurface Characterization, Permeability |
| Key Technologies | CNNs, PINNs, GANs, PhaseNet, Python, TensorFlow, PyTorch |
| Eligibility | Ph.D. in Geophysics, Geology, Petroleum Engineering, Computer Science, or related field |
| Ph.D. Requirement | Degree completed within five years of the appointment start date |
| Citizenship | United States citizenship required |
| Application | University Talent Center + email application materials |
| Job ID | 26005201 |
| Last Date for Apply | Not specified in the provided announcement |
Designation
Postdoctoral Researcher
Research Area
- Machine Learning and Deep Learning
- Artificial Intelligence for Subsurface Applications
- Geophysics and Geology
- Petrophysics and Rock Physics
- Subsurface Multiscale Structure Characterization
- Permeability Prediction
- Ultrasonic Acoustic Measurements
- Seismic Data Analysis
- Scientific Machine Learning
- Physics-Informed Neural Networks
- Generative AI and Synthetic Data
Location
University of Pittsburgh, Pittsburgh, Pennsylvania, USA
Campus: Pittsburgh
Eligibility/Qualification
Required Qualifications
- Ph.D. in Geophysics, Geology, Petroleum Engineering, Computer Science, or a closely related field.
- The Ph.D. must have been completed within the last five years from the start date of the appointment.
- United States citizenship is required.
- Demonstrated experience applying machine learning or deep learning techniques to scientific problems.
- Strong proficiency in Python and experience with common machine-learning/deep-learning libraries such as TensorFlow or PyTorch.
- Strong analytical and problem-solving abilities.
- Excellent written and oral communication skills.
- Ability to work independently as well as collaboratively within a multidisciplinary research team.
Preferred Qualifications
- Experience working with geophysical, petrophysical, or well-log datasets.
- Background in rock physics, acoustics, or seismic data analysis.
- Experience with advanced neural-network architectures, including CNNs, PINNs, or GANs.
- Strong scholarly record, including first-author publications in peer-reviewed journals.
- Familiarity with high-performance computing (HPC) environments.
- Experience developing relational databases and database schemas.
Job Description
The successful candidate will play a central role in developing an AI/ML-based approach for subsurface characterization and permeability prediction.
Key responsibilities include:
- Construct a fully attributed, machine-learning-ready petrophysical database using existing NETL ultrasonic and core measurement archives.
- Develop, train, and deploy deep-learning models, including convolutional neural networks (CNNs) and physics-informed neural networks (PINNs), to predict rock permeability from ultrasonic acoustic measurements.
- Adapt and retrain existing deep-learning frameworks such as PhaseNet to automate P- and S-wave arrival picking from ultrasonic waveform data.
- Develop and apply generative adversarial networks (GANs) to generate realistic synthetic core data and expand training datasets.
- Integrate and validate AI/ML models using existing wireline log data and potentially new core samples.
- Collaborate with scientists and researchers at NETL and experts in geophysics, geology, engineering, and computer science.
- Publish research results in high-impact, peer-reviewed journals.
- Present research findings at major scientific conferences.
- Contribute to the development of innovative AI/ML tools for subsurface energy-resource characterization, production efficiency, and recovery optimization.
How to Apply
Applicants should apply through the University of Pittsburgh Talent Center job posting for Job ID 26005201.
In addition, applicants should email Dr. William Harbert at harbert@pitt.edu with the subject line:
“Postdoc Opportunity”
The following documents should be combined into one PDF attachment:
- Cover letter describing career goals, relevant experience, and interest in the position.
- Resume/CV including contact information for three professional references.
Required Attachments: Cover Letter and Curriculum Vitae
The University of Pittsburgh is an equal opportunity employer, including for individuals with disabilities and veterans.
Last Date for Apply
Open until filled. Applicants are advised to check the University of Pittsburgh Talent Center posting for the current application deadline and posting status.
