Seismic ML Scientist: Inversion & Imaging

Occidental Petroleum Corporation; United States; On-site; USD 140,000 - 200,000; Full time

Job summary

Occidental Petroleum Corporation seeks a Machine Learning Expert for its Subsurface Innovation Imaging Team in Houston, TX. The role focuses on developing ML and hybrid physics-ML methods for seismic imaging, inversion, and velocity-model building, including ML-assisted RTM, LSRTM, and FWI.

Applicants should have 5+ years post-PhD experience with FWI on real data, strong Python and PyTorch skills, and a track record applying ML to physics-based problems.

Qualifications

• MS/PhD in Geophysics, Applied Mathematics, Physics, Computer Science, or a related quantitative field.
• Minimum 5 years of hands-on post-PhD FWI experience on real seismic data.
• Fluency with ML applied to geophysical or physics-based problems.
• Strong Python and PyTorch skills; experience training/validating models.

Responsibilities

• Develop ML and hybrid physics-ML methods for seismic imaging, inversion, and velocity model building including ML-assisted RTM, LSRTM, and FWI.
• Apply deep learning, generative models, neural operators, and physics-informed approaches to denoise, interpolate, enhance images, predict models, quantify uncertainty, and interpret data.
• Build workflows to handle large multidimensional seismic datasets on GPU and cloud infrastructure.
• Validate new methods against physics-based approaches and assess value vs. limitations.
• Move prototypes into production imaging workflows with proper documentation and reproducibility.
• Communicate results clearly to geophysicists, data scientists, and asset teams.

Skills

Python

PyTorch

Geophysics

FWI experience

GPU computing

Education

MS/PhD in Geophysics/related field

Tools

AWS

HPC

GPU Cluster

Job description

Occidental Petroleum Corporation seeks a Machine Learning Expert for its Subsurface Innovation Imaging Team in Houston, TX. The role focuses on developing ML and hybrid physics-ML methods for seismic imaging, inversion, and velocity-model building, including ML-assisted RTM, LSRTM, and FWI.

Applicants should have 5+ years post-PhD experience with FWI on real data, strong Python and PyTorch skills, and a track record applying ML to physics-based problems.