Ride-Share Surge Predictor Workflow

This n8n workflow is designed to predict ride-share surge conditions by analyzing real-time data and generating insights through AI models. It uses webhooks to trigger the process, splits incoming text data into manageable parts, creates embeddings for each part using Cohere, and stores these vectors in a Supabase vector database. The system then queries the database to retrieve relevant information, processes queries with a language model, and maintains memory buffers for context. The workflow also employs an agent to define insights, with final results logged into a Google Sheet for record-keeping. This automation is highly useful for ride-sharing companies or analysts seeking to anticipate surge times based on data patterns, enabling more informed decision-making and dynamic pricing strategies.