This project introduces an innovative AI model tailored for automated menu review in kitchens, powered by the custom-trained Llama2 architecture and enriched with training data generated using Claude AI. Developed to address the scarcity of available training data, our approach ensures a focused and effective solution for menu evaluation tasks. By harnessing the capabilities of the custom-trained Llama2 model, our system performs comprehensive menu assessments, evaluating factors such as dish uniqueness, ingredient availability, pricing, and regulatory compliance. Additionally, the integration of data generated by Claude AI enhances the model's ability to accurately assess menus by providing tailored training data for improved performance in real-world scenarios. Through the synergy of the custom-trained Llama2 architecture and Claude AI-generated training data, our system facilitates efficient and accurate menu review processes. By effectively identifying discrepancies and issues, it streamlines manual review efforts, contributing to enhanced operational efficiency and quality control in kitchen operations.
