Human–AI Teaming and Agentic Reasoning for Cyber-Physical Social Systems in Manufacturing

Overview

As part of its AI for Manufacturing project, NIST develops the measurement science — methodologies,
metrics, datasets, tools, and benchmark studies — needed to enable trustworthy AI and reliable human-
AI teaming across manufacturing system design, operation, and automation. A critical research thrust
concerns human-AI teaming and the emergence of Cyber-Physical Social Systems (CPSS), in which
humans and AI agents sense, reason, and act together within bounded, verifiable autonomy loops. This
direction anticipates an Industry 5.0 paradigm in which human and machine intelligence increasingly
converge, and in which effective collaboration must extend beyond conventional human-in-the-loop and
human-on-the-loop arrangements. The Research Associate will investigate advanced human-AI
relationship models, the integration of complementary reasoning modes for manufacturing operations,
and the systematic engineering of prompts and agent skills, and will develop systems-engineering
approaches to specify, design, test, and validate CPSS for manufacturing. The work draws on
neurosymbolic reasoning, reinforcement learning, large language model and agentic-AI methods, formal
modeling, and systems engineering, applied to representative manufacturing use cases.

Qualifications

• A PhD degree (completed or expected) in Computer Science, Systems Engineering, Industrial Engineering, or a related field.
• Research experience in human-AI teaming, cyber-physical (social) systems, or intelligent autonomous systems.
• Familiarity with neurosymbolic architectures, reinforcement learning, and large language model and agentic-AI methods, including retrieval-augmented generation (RAG) and Model Context Protocol (MCP).
• Experience with systems-engineering methods for specification, design, verification, and validation, such as model-based systems engineering (e.g., SysML) and formal models (e.g., Petri nets).
• Familiarity with manufacturing systems and operations is desirable.
• Ability to develop prototypes of tools and models needed to conduct and evaluate the research.
• Strong oral and written communication skills and a record of research publication.
• U.S. Citizen Preferred.

Research Proposal

Key responsibilities will include but are not limited to:

• Investigating human-AI relationship models for manufacturing that extend beyond human-inthe- loop and human-on-the-loop collaboration — including AI-as-mentor, AI-as-coach, and AI-aschallenger configurations and their reciprocal forms — within cyber-physical social systems, and examining the associated ethical dimensions.
• Developing systems-engineering approaches and methods to specify, design, test, and validate cyber-physical social systems for manufacturing, including metrics for the quality, trustworthiness, and safety of human-AI teaming under operational conditions.
• Investigating how complementary reasoning modes — physics-based and causal models, human knowledge represented in prompts and agent skills, Model Context Protocol (MCP) tool descriptions, and generative-AI pattern-recognition reasoning — can be integrated to support manufacturing system operations.
• Developing a systematic methodology for engineering prompts, system prompts, and agent skills for manufacturing engineering and operations, grounded in an agent/AI function and role reference model, with potential specialization to discrete, batch, or continuous manufacturing, or to a specific sector such as semiconductor manufacturing.
• Presenting results at internal meetings and occasional meetings with external stakeholders, and contributing to publications and relevant standards-development efforts.