The aim of this research is to inform how Human-Machine Teams (HMTs) engaged in multi-domain operations can build and maintain trust in synthetic teammates driven by complex Artificial Intelligence (AI) to improve mission outcomes. Warfighter performance will depend increasingly more on mission-specific tasks performed by unmanned vehicles with sophisticated AI to aid in team tasks and decision making. To be effective, these robotic teammates need to behave much like humans in terms of transparent decision-making and communications. Thus, it is imperative to understand (a) how robotic teammates should interact with humans as an intact, cohesive team, and (b) how these HMTs build and sustain trust. Our approach investigates trust and how it can be measured in real-time between humans and robotic teammates. In our paradigm, robotic teammates are driven by the Ontogenesis Engine for Man-Unmanned Systems (OEMUS), which is composed of interconnected AI capabilities designed to support a teams’ Observe-Orient-Decide-Act (OODA) loop. In addition, biometric sensors are used to glean psychophysiological data to train real-time Machine Learning (ML) trust classifiers. Level of trust then mediates HMT coordination via multimodal Human-Machine Interface (HMI) designed around HMT goals that promote shared situation awareness (SA). Shared SA is obtained by displaying pertinent team decisions and suggestions, and real time trust indices are used to moderate information presented on these shared displays. Trust is built and sustained over time by keeping a record of HMT interactions and mediating level of information presented based on that history. Theoretical findings are presented in terms of domain-specific guidelines, principles and requirements necessary to (a) measure trust based on biometric sensor data, and (b) build and sustain trust via robust HMI interfaces.