In the context of the BrainSoup platform, the term Working Memory denotes the context window of a Large Language Model (LLM). In operation, the Working Memory holds essential information that instructs the model's responses, thereby personalizing interactions with users. The data size injected into these context windows is consequential as it affects both the performance and cost-efficiency of the LLM. Specifically, the size of the data can impact the computational resources required for the model's operation, which in turn affects the LLM usage costs. As such, it is recommended to optimize data input to ensure maximum efficiency.