Postdoc: Generative AI, Multi-Agent Safety & Operational Tools

UT-Battelle; Oak Ridge (TN); On-site; USD 65,000 - 90,000; Full time

Benefits offered by this job

Medical benefits

Relocation assistance

Flexible work hours

Job summary

UT-Battelle is seeking a Postdoctoral Research Associate to lead research at the frontier of Generative AI, LLMs, and multi‑agent AI workflows for high‑consequence operations at ORNL. The role emphasizes practical tools for predictive safety, risk assessment, and incident prevention, with access to large‑scale GPU resources and opportunities to publish.

You will develop, fine‑tune, and evaluate AI architectures, collaborate across CSMD and operations teams, and contribute to trusted RAG systems

Qualifications

• Ph.D. in CS/AI/EE/Applied Math or related field; candidates may finish by end of 2026.
• Strong ML/DL background with practical experience in generative AI, multi‑agent AI systems, LLM fine‑tuning.
• Proficient Python and ML libraries (Hugging Face, LangChain, scikit‑learn, pandas).
• Publications or proven technical tool development in AI/ML/NLP.

Responsibilities

• Conduct original research and software tool development in Generative AI for operational safety and risk mitigation.
• Architect and deploy multi‑agent AI systems and RAG pipelines for automated risk scoring and hazard reporting.
• Develop validation frameworks, explainable AI methods, and safety‑critical alignment techniques.
• Collaborate with cross‑disciplinary teams to integrate AI tools into safety management systems.
• Publish in AI/ML and safety journals; contribute to proposals and open‑source releases.

Skills

Machine learning

Deep learning

Python

PyTorch

TensorFlow

Hugging Face

LangChain

LlamaIndex

Research

Education

PhD in Computer Science / AI / EE / Applied Math or related, expected by end of 2026

Tools

GPU clusters

Job description

UT-Battelle is seeking a Postdoctoral Research Associate to lead research at the frontier of Generative AI, LLMs, and multi‑agent AI workflows for high‑consequence operations at ORNL. The role emphasizes practical tools for predictive safety, risk assessment, and incident prevention, with access to large‑scale GPU resources and opportunities to publish.

You will develop, fine‑tune, and evaluate AI architectures, collaborate across CSMD and operations teams, and contribute to trusted RAG systems