This document summarizes a presentation about using recurrent neural networks like LSTMs and AWS Lambda for real-time energy price forecasting and optimization. The objectives are to provide a cost-effective real-time market price forecasting service and scalable distributed solution. LSTMs are well-suited for this task as they can learn nonlinear relationships in time series data and remember long-term dependencies. The architecture proposed uses AWS Lambda to run prediction code in response to new time series data and store results in S3.