Brand Bloom is an innovative recursive method designed to generate the best brand names based on input parameters. This README will guide you through the workflow and key steps involved in using Brand Bloom effectively.

Brand Bloom operates through a series of well-defined steps to generate high-quality brand names:

Begin by customizing your query to meet your specific requirements. This query serves as the foundation for generating brand names that align with your project or business.

The query is sent through a query expansion process. During this step, the query is enriched with additional information, details, and elements that might not have been initially considered. This expansion ensures that the generated brand names are comprehensive and well-informed.

After query expansion, the expanded query is parsed into a structured object. This structured representation serves as the input for generating potential brand names.

Brand Bloom utilizes Language Model (LLM) technology to generate a list of potential brand names. These names are creative, memorable, and aligned with the information in the expanded query.

Each generated brand name is subjected to a scoring process. The scoring is based on various criteria, and if a brand name doesn't meet the predetermined score threshold (e.g., 0.9), further refinement is required.

When a brand name falls below the desired score threshold, Brand Bloom provides a critique of the last generated options. This critique offers valuable guidance on how to enhance the brand names and improve their quality.

The entire process operates in a recursive loop until at least one brand name is found that meets or exceeds the defined score threshold. This iterative approach ensures that you obtain the best possible brand names for your project or business.