Differential privacy (DP) has garnered significant attention from both academia and industry due to its potential in offering robust privacy protection for individual data during analysis. With the increasing volume of sensitive information being collected by organizations and analyzed through SQL queries, the development of a general-purpose query engine that is capable of supporting a broad range of SQLs while maintaining DP has become the holy grail in privacy-preserving query release. However, there are two significant challenges. First, guaranteeing privacy in a relational database with multiple relations, foreign keys, and the join operator is challenging since individuals can make large and correlated contributions to the query results. Second, noise injection is essential for privacy protection, but traditional notions of optimality, such as instance optimality and worst-case optimality, are either unachievable or meaningless when evaluating relational queries under DP, further complicating the task of achieving an optimal privacy-utility trade-off. In this talk, I will give a selective overview of my recent research in addressing these challenges in SQL queries answering under DP.