Generally, side slotting and directional techniques can improve the performance of a conventional Vivaldi antenna (CVA), but the optimal structure and distribution of slots and directors may be irregular or even complex, requiring significant manual effort, thus limiting the design possibilities. In this paper, a genetic algorithm (GA) is introduced to assist in designing and optimizing a new type of balanced side slotted Vivaldi antenna with director (SSVAD). The methods of artificial intelligence make the process of searching for the optimal structure of such a multi-objective and multi-dimensional problem simpler and more diverse. The GA-generated SSVAD antenna consists of a CVA and 34 slots with varying lengths, as well as 5 metal strips. It has a compact size of 38.2×49×0.8 mm3 (or 0.32λL×0.41λL×0.007λL, where λL is the lowest operating frequency of 2.48 GHz). The measured results show that the antenna has a peak gain >0 dBi over 2.48-10.88 GHz and >5 dBi over 4.6-10.88 GHz with S11<-10 dB standard, and exhibits directional characteristics at most of the operating frequencies. Since the measured results are basically consistent with the simulation ones, the effectiveness of the designed scheme has been proven.