Applied Data Scientist - Staff Member

Oak Ridge National Laboratory•Oak Ridge, TN

1d•Onsite

About The Position

We are seeking an Applied Data Scientist to perform development of novel and advanced AI/ML algorithms. This work will support a broad user base by applying AI methods that span low-power edge computing utilizing spiking neural network algorithms to large models utilizing agentic AI to solve complicated problems in nuclear safeguards, warhead monitoring and fundamental physics. Experience with sensor modalities such as radiation detectors, event-based cameras and seismo-acoustic sensors is needed. This role resides in the Advanced Detection and Applied Data Science (ADADS) group, Nuclear Structure & Advanced Technologies Section, Physics Division, Physical Sciences Directorate, at Oak Ridge National Laboratory (ORNL). ADADS draws from fundamental physics to develop technologies and processes that support the design, development, and deployment of high-sensitivity detection systems for measuring radiation and other physics phenomenologies. The group focuses on the development and evaluation of novel and advanced radiation detector and non-destructive evaluation concepts for basic science, nonproliferation, national security, and intelligence organizations. The ADADS group performs the research and technology development to evolve these advanced concepts from conception to demonstration or deployment with particular emphasis on sensitivity and specificity in actual field operations, including use of radiation transport codes to model these systems and analysis codes to isolate signals of interest.

Requirements

Bachelor’s degree in physics, computer science, mathematics, or a related field
A minimum of 5 years of experience, post bachelor’s, demonstrating capabilities and experience in data science or data analytics
Ability to obtain and maintain a clearance from the Department of Energy

Nice To Haves

Basic understanding of the detection of radioactive materials through measurement of ionizing radiation
Experience with applications in nuclear non-proliferation and fundamental physics
Experience in one or more of the following ML research areas: Neuromorphic computing, Uncertainty quantification, Unsupervised/Semi-supervised learning and data mining techniques (clustering, embeddings, dimensionality reduction, anomaly detection), Physics informed machine learning
Practical experience using ML models: Dataset curation and preprocessing, Applications leveraging off-the-shelf ML models, Fine-tuning pre-trained models for specific tasks, Designing and training models, from scratch
Familiarity with relevant technologies such as: Computer programming languages such as Python, JavaScript, FORTRAN, C, and C++, Machine learning libraries such as Pytorch, TensorFlow, Scikit-Learn, and OpenCV, 3D game development software such as Unity including VR/AR packages
Commitment to stay current with modern data analytics/machine learning trends
A strong interest in problem solving and applying new skills and methods
Excellent human relation and oral and written communication skills and a demonstrated ability to work in a team-oriented environment with a broad range of domestic or international collaborators
Ability to work both independently, with minimal supervision, or as an effective member of an agile development team

Responsibilities

Develop and apply modern data science/machine learning techniques to a wide variety of subjects including the characterization of nuclear material and nuclear facility operations, as well as more fundamental physics research.
Develop and test novel data analytics processes and specialized methods to improve measurement fidelity and reduce uncertainties in models used for real-world decisions.
Document results by publication in high impact papers, journals, conference papers and technical reports.
Collaborate in a team environment performing project work and developing research proposals.