Studies of human factors in aviation by the Flight Safety Foundation have strongly associated chronic and acute stress with pilot error. Our team's pilot health monitoring system aims to mitigate risks associated with pilot stress by identifying when a pilot is experiencing stress and notifying other avionics systems in the cockpit. Our team initially analyzed existing consumer-grade wearable bio-metric devices to determine the feasibility of the project. We identified common strengths, weaknesses, and technologies shared among the devices. This heavily influenced our initial design: a wearable device that measures heart rate and blood oxygen saturation, then communicates wirelessly with a base station. This enabled another ECE capstone hardware team to create a lightweight/comfortable wearable while not compromising on the compute required for machine learning. Our team focused on the software. In C++, we developed a low-latency system that saves, streams, and classifies pilot data in real-time. We also created a dashboard that can be accessed wirelessly using an Electronic Flight Bag (EFB) device. As a team, we learned a lot about collaborative software projects and how to design good system architecture. We encountered numerous challenges including difficulties with integrating software components and acquiring data to train and evaluate the efficacy of our logistic regression classifier. Although the project did not achieve some of our stretch goals/objectives, we are proud of the software we created! https://eecs.oregonstate.edu/capstone/submission/pages/viewSingleProject.php?id=jbS996Tu2DNSHhhX