Build a product framework around real time log analysis, reporting and alerting using Amazon AWS.
The client’s existing system was studied and requirements to improve the relevance of ad placement and to provide a fail-safe platform for advertisers & ad managers to view the consumption and effectiveness of the ads were documented. Revenue impact of deploying a improvised solution was studied and the cost benefit analysis were presented to the client. A server-less big data cloud solution architecture was put together to handle and analyse large volume of ad trackers on YouTube video channels for Roku &FireTV. AWS components were picked to handle high volume transactional data handling and to optimize operational costs. Statistical models were built against historical data to study ad effectiveness and revenue generation. Models were built to consume the improvised data features of the new system being built and integrated to the ad serving engine.