Vladimir Kazei

News

• Joined ExxonMobil as a Geophysics Specialist (Spring, TX).
  November 2026
• New preprint: Geometry of Confabulation and Refusal in Small Language Models (arXiv:2611.01272)
  October 31, 2026 · Independent research, done in a personal capacity · Code and data
• Advancing Data Analytics & Machine Learning for Exploration Geophysics
  SEG Workshop · Co-organizer & speaker · May 18–21, 2026 · Houston, TX
• Post-convention workshop: Agentic AI and Large Language Models (LLMs) in Geophysical Applications
  IMAGE 2026 · Lead organizer · August 2026 · Houston, TX
• Call for papers: Machine Learning — special section in The Leading Edge
  Publication July 2026 · Special-section coordinator · Guest editors: O. Ovcharenko, H. Di, S. Farris, Q. Zhang
• Special section published: Generative and Physics-Informed AI
  The Leading Edge 44(2), 2025 · Special-section coordinator
• Seismic Soundoff: Generative, Physics-Informed, and Agentic AI in Geophysics
  SEG Podcast · April 2025

About

I am a Geophysics Specialist – Advanced at ExxonMobil (since November 2026). Previously Geophysicist (Senior from 2023) at Aramco Americas Houston Research Center (November 2020–October 2026), working on machine learning and distributed acoustic sensing (DAS): DAS-based VSP imaging, deep learning-driven full-waveform inversion (FWI), and generative AI for subsurface modeling.

Associate Editor of Geophysics (2025–2027). Previously Research Scientist at KAUST in the Seismic Wave Analysis Group. PhD in Geophysics from Saint Petersburg State University.

Selected Publications

• Introduction to this special section: Generative and physics-informed AI
  V. Kazei, O. Ovcharenko, H. Di & U. B. Waheed · The Leading Edge 44(2), 2025
• Diffusion model for DAS-VSP data denoising
  D. Zhu, L. Fu, V. Kazei & W. Li · Sensors 23(20), 2023
• Distributed acoustic sensing in geophysics — Introduction
  Geophysics 88(6), 2023 · Special section introduction, co-editor
• Inverting distributed acoustic sensing data using energy conservation principles
  V. Kazei & K. Osypov · Interpretation, 2021
• Mapping full seismic waveforms to vertical velocity profiles by deep learning
  V. Kazei, O. Ovcharenko, P. Plotnitskii, D. Peter, X. Zhang & T. Alkhalifah · Geophysics, 2021
• Scattering radiation pattern atlas: What anisotropic elastic properties can body waves resolve?
  V. Kazei & T. Alkhalifah · Journal of Geophysical Research: Solid Earth, 2019
• Deep learning for low-frequency extrapolation from multi-offset seismic data
  O. Ovcharenko, V. Kazei, M. Kalita, D. Peter & T. Alkhalifah · Geophysics, 2019

Full list on Google Scholar.

Code

• MedSeis Medical imaging startup adapting seismic methods to ultrasound (co-founder)
• ScatteringAtlas Radiation patterns of elastic anisotropic parameters in the wavenumber domain
• deeplogs Convolutional neural network that builds velocity models from raw seismic data
• fastHelmholtz Fast 2D Helmholtz solver with absorbing boundaries for frequency-domain FWI
• fastFWI Simple full-waveform inversion code to test regularizations, built on fastHelmholtz