With the rapid development of smart campuses, traditional campus second-hand trading methods can no longer meet the growing needs of students. This paper proposes a design scheme for an intelligent campus second-hand trading platform based on microservices and cloud computing, which provides customized trading experience for users by introducing advanced recommendation algorithms. The microservice architecture gives the platform extremely high module decoupling and autonomous scalability, ensuring the high availability and agile response of the system. The recommendation system uses a hybrid algorithm, combining content filtering and collaborative filtering, and continuously optimizes the recommendation results through machine learning, so as to accurately capture user preferences. One of the features of the platform is its community-driven data feedback mechanism, which not only can adjust the recommendation strategy in real-time but also can promote the sustainable development of campus economy and culture through data mining. In addition, the research explores a novel credit evaluation system design to establish a trustworthy trading environment. The experimental results show the technical innovation and high efficiency of the platform in multiple dimensions, which is significantly superior to the existing solutions. The successful implementation of this research indicates that the campus trading platform is moving towards a more intelligent and personalized direction, and is expected to play a key role in the smart campus ecosystem.