Nuclear Power Applications Postdoctoral Data Science Researcher

Idaho National Laboratory • Idaho Falls, ID

7d • Onsite

About The Position

The Idaho National Laboratory (INL) is seeking a well-qualified Postdoctoral Research Associate in Data Science for the Department of Plant Optimization. The ideal candidate should possess a robust technical background and practical experience with large language models (LLMs). The Postdoctoral Researcher will have the following key responsibilities: Develop, fine-tune, and optimize LLMs. Design and implement efficient retrieval-augmented generation systems. Create advanced prompt engineering strategies to enhance model performance and utility. Demonstrate expertise in embedding techniques, with a clear understanding of their strengths, weaknesses, and applicability to various use cases. Perform comprehensive data preprocessing, including cleaning, normalization, and integration of diverse data types such as text, images, and tabular data. Analyze and structure complex datasets, including unstructured data types such as images and tables. Apply advanced AI model optimization techniques to improve the efficiency and performance of large and frontier models for specific applications. Understand the principles and trade-offs associated with large-scale machine learning systems. Develop agentic AI systems including multi-modal models. Communicate effectively, both verbally and in writing, with management, co-workers, and customers. Publish research results in a mix of journal publications and conference proceedings. Write or contribute to the writing of technical reports. Prepare competitive proposals for the DOE or internal INL funding opportunities. Give technical presentations to internal and external audiences.

Requirements

• PhD in Data Science, Engineering, Mathematics, Computer Science, Artificial Intelligence or comparable disciplines.
• PhD requirements must be completed by commencement of appointment and within the previous 5 years.
• Experience developing and optimizing large language models (LLMs).
• Experience in data science, data analysis, and software development using Python, C#, Java or C++.
• Demonstrated oral and written communication in English.
• Must be either a US Citizen or a US Green Card holder to qualify for the job.

Nice To Haves

• Experience in distributed processing frameworks and high-performance computing resources to build and deploy scalable AI solutions.
• Experience in model optimization and deployment on NVIDIA and AMD GPUs, ensuring efficient utilization of computational resources.
• Experience in computer vision or time-series data analytics, including practical applications in real-world scenarios.
• Experience with vision language models and large behavioral models.
• Knowledge of data architecture, data structuring, and best practices for managing large-scale data pipelines.
• Familiarity with nuclear engineering concepts, including nuclear reactors and associated data, is a significant advantage.
• Ability to work independently and collaboratively in a multidisciplinary team environment.

Responsibilities

• Develop, fine-tune, and optimize LLMs.
• Design and implement efficient retrieval-augmented generation systems.
• Create advanced prompt engineering strategies to enhance model performance and utility.
• Demonstrate expertise in embedding techniques, with a clear understanding of their strengths, weaknesses, and applicability to various use cases.
• Perform comprehensive data preprocessing, including cleaning, normalization, and integration of diverse data types such as text, images, and tabular data.
• Analyze and structure complex datasets, including unstructured data types such as images and tables.
• Apply advanced AI model optimization techniques to improve the efficiency and performance of large and frontier models for specific applications.
• Understand the principles and trade-offs associated with large-scale machine learning systems.
• Develop agentic AI systems including multi-modal models.
• Communicate effectively, both verbally and in writing, with management, co-workers, and customers.
• Publish research results in a mix of journal publications and conference proceedings.
• Write or contribute to the writing of technical reports.
• Prepare competitive proposals for the DOE or internal INL funding opportunities.
• Give technical presentations to internal and external audiences.

Benefits

• Medical, Dental, Vision, and Flexible Spending Accounts
• 401(k) with a 4.2% employer contribution and up to 4.8% match (regular positions) or self-contribute access (postdoctoral positions)
• Paid time off (personal leave)
• Employee Education Program (tuition assistance for eligible positions)
• Comprehensive Relocation Package

What This Job Offers

• Job Type: Full-time
• Career Level: Entry Level
• Education Level: Ph.D. or professional degree