media-search is a service for searching for media such as on YouTube by queries, accessible both on the commandline and on a web user interface, via HTTP requests. It is instrumented with OpenCensus which gives us distributed tracing and monitoring for visibility into our entire system as a call propagates through the various microservices. Normal access is by hitting the frontend service (OFE) via HTTP requests to /search with either GET or POST methods. OFE then invokes a gRPC transport based connection to the search backend(SB) which then searches for content from YouTube. Because content search from YouTube incurs API quota costs as well as time expense having to fetch from YouTube's servers, it helps to cache results so that subsequent repetitions will return cached content in very little time. Once results have been returned during a cache-miss, they are cached to MongoDB and a subsequent asynchronous call is made to a gRPC accessible service that then fetches individual meta information about each video and also caches that to MongoDB.