Postdoctoral Researcher - Explainable AI for 3D Data

At ExxonMobil, we apply advanced optimization and machine learning techniques to solve some of the most challenging problems in energy, manufacturing, and low-carbon technologies. In this role, you will work on cutting-edge methods at the intersection of OR and AI, directly impacting critical business decisions and shaping next-generation computational decision-support capabilities.

ExxonMobil is seeking a highly motivated Postdoctoral Researcher specializing in Explainable Artificial Intelligence (XAI) for large-scale 3D data analysis. The successful candidate will develop interpretable machine learning methods for segmentation, classification, and anomaly detection in high-dimensional volumetric datasets to support critical business and engineering decisions.

This role is ideal for a recent Ph.D. graduate with expertise in XAI and deep learning applied to complex spatial data. The candidate will work closely with domain experts to create transparent, trustworthy AI systems that provide actionable insights for high-stakes applications.

Key Responsibilities

Develop explainable AI methods for deep learning models applied to 3D volumetric data.

Design and implement models for segmentation, classification, and anomaly detection in large-scale datasets.

Create techniques to improve model interpretability, transparency, and trustworthiness, including post hoc explanation and inherently interpretable approaches.

Develop uncertainty-aware predictions to support decision-making in critical applications.

Optimize models for scalability and performance on large 3D datasets.

Evaluate models using both predictive accuracy and explainability metrics relevant to domain needs.

Collaborate with domain experts to translate model outputs into decision-support tools.

Implement workflows using modern ML frameworks and reproducible software practices.

Communicate findings through technical reports, journal publications, and conference presentations.

Explainable AI methods for deep neural networks

3D computer vision and volumetric data analysis

Semantic and instance segmentation in large 3D volumes

Anomaly detection in high-dimensional spatial data

Interpretable representations for classification models

Uncertainty quantification and confidence estimation in AI models

Human-in-the-loop AI and decision-support systems

Applications to subsurface imaging, industrial inspection, and sensor data

Ph.D. in Computer Science, Artificial Intelligence, Machine Learning, Computational Science, or a closely related field, with a focus on explainable AI or interpretable machine learning.

Demonstrated research experience in explainable AI and deep learning, including one or more of:

Model interpretability (e.g., saliency methods, attribution, feature importance)

Explainability techniques for neural networks

Interpretable model design

Experience with 3D data (e.g., volumetric imaging, point clouds, or spatiotemporal data) and deep learning methods such as CNNs, transformers, or graph neural networks.

Proven experience in segmentation, classification, or anomaly detection tasks.

Strong programming skills in Python.

Hands-on experience with machine learning frameworks such as PyTorch or TensorFlow.

Strong analytical, problem-solving, and communication skills.

Ability to work effectively in multidisciplinary teams.

Experience with XAI methods for computer vision or 3D data.

Familiarity with uncertainty quantification, probabilistic ML, or Bayesian deep learning.

Experience with large-scale data processing and GPU-accelerated training.

Knowledge of evaluation metrics for explainability and model trustworthiness.

Experience applying AI to engineering, geospatial, industrial, or scientific datasets.

Strong publication record in XAI, machine learning, or computer vision.

Demonstrated ability to translate research into decision-support applications.

Passion for developing trustworthy and interpretable AI systems.

Interest in solving high-impact, real-world problems involving complex data.

Ability to bridge machine learning methods with practical decision-making needs.

Collaborative mindset and strong communication skills.

Self-driven with the ability to independently lead research initiatives.

This opportunity is for a postdoctoral position expected to last one to three years, subject to annual review and renewal.

An ExxonMobil career is one designed to last. Our commitment to you runs deep: our employees grow personally and professionally, with benefits built on our core categories of health, security, finance, and life. Individual pay is determined based on various factors including degree/education, discipline, year of study, skills, abilities, qualifications, and work experience.

More information on our Company’s benefits can be found at www.exxonmobilfamily.com.

Please note pay rates and benefits may be changed from time to time without notice, subject to applicable law.

Relocation benefits may be available to you based on ExxonMobil eligibility guidelines.

ExxonMobil is an Equal Opportunity Employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, age, sexual orientation, gender identity, national origin, citizenship status, protected veteran status, genetic information, or physical or mental disability.

Nothing herein is intended to override the corporate separateness of local entities. Working relationships discussed herein do not necessarily represent a reporting connection, but may reflect a functional guidance, stewardship, or service relationship.

Exxon Mobil Corporation has numerous affiliates, many with names that include ExxonMobil, Exxon, Esso and Mobil. For convenience and simplicity, those terms and terms like corporation, company, our, we and its are sometimes used as abbreviated references to specific affiliates or affiliate groups. Abbreviated references describing global or regional operational organizations and global or regional business lines are also sometimes used for convenience and simplicity. Similarly, ExxonMobil has business relationships with thousands of customers, suppliers, governments, and others. For convenience and simplicity, words like venture, joint venture, partnership, co-venturer, and partner are used to indicate business relationships involving common activities and interests, and those words may not indicate precise legal relationships.