In this study, we investigated the application of large language models (LLMs), a subset of AI systems specifically designed for natural language understanding and generation, in undergraduate electronics design competitions, with a focus on how these AI tools can facilitate human–AI collaboration. Using the National Undergraduate Electronics Design Contest as a case study, we examined how LLMs can enhance the design process, improve problem-solving skills, and facilitate the iterative optimization of engineering solutions. The design task centers on an ultrasonic audio jamming system (commonly referred to as a “recording shielding system”) that integrates key sensor-related technologies, including ultrasonic transducers, microphone-based audio monitoring, and frequency-domain signal analysis using the fast Fourier transform. These components form a functional sensor application scenario, where the system detects and discriminates audio signals to enable a dynamic response. In this study, we examined the application of LLM-driven tools to assist students during the design and development of the sensor-based system, demonstrating how these tools supported students in addressing complex technical challenges. We highlight the advantages of using AI for in-design proposal generation, system optimization, and troubleshooting, while also addressing the challenges of ensuring that AI-generated solutions are accurate and feasible within competition constraints. The findings suggest that LLMs make a significant contribution to students’ learning outcomes, fostering creativity, critical thinking, and technical proficiency in real-world engineering contexts. In this paper, we offer valuable insights into the evolving role of AI in enhancing educational practices in engineering design.