Design and scale a generalizable hybrid physics-AI foundation model on real seismic datasets.
Propose advanced techniques including AI-assisted velocity model building, learned regularization and preconditioning, cycle-skipping mitigation, and accelerated forward modeling/gradient computation.
Scale implementations for GPU clusters, HPC systems, and large-scale 3D datasets.
Integrate AI into FWI and imaging pipelines
Seamlessly embed AI components into existing Full Waveform Inversion and seismic imaging workflows.
Promote best practices in scientific software development and ML lifecycle management.
Validate, deploy, and drive business impact
Validate solutions on field data in complex geological settings and establish clear performance, robustness, and risk metrics.
Collaborate with stakeholders to ensure geological accuracy and commercial relevance.
Mentor the team on FWI fundamentals while driving adoption of modern AI/ML techniques.
Required Qualifications
PhD/MS in Geophysics, Applied Math, Physics, Computer Science, or equivalent experience
Deep expertise in Full Waveform Inversion, seismic wave propagation, and inverse problems
Prior hands-on industry experience in subsurface imaging is a plus
Track record of developing and delivering production-grade scientific software in HPC or cloud environments, including data pipeline and imaging algorithm
Preferred Qualifications
Expertise in physics-informed machine learning, learned solvers, or neural surrogate models.
Proven success applying deep learning to physics-based modeling and simulation
Experience deploying ML models at scale (lifecycle management, monitoring, reproducibility)
Track record of high‑impact publications, patents, or industrial innovations
Strong Python programming skills with hands-on experience in PyTorch or JAX