openzipkin/zipkin ? reverse-engineered prompt

Build me a distributed tracing app like Zipkin, with a simple web UI where I can search and view traces by trace ID, service name, operation name, tags, and duration. I want to be able to paste a trace ID from logs and jump straight to that request, then see timing details, failures, and a dependency view of how services call each other.

It should support receiving span data over HTTP, and be set up so other tracing libraries can send data into it. Keep the first version working with in memory storage, but structure it so persistent storage like Elasticsearch or Cassandra could be plugged in later. Make it easy to run locally with a single command, and include a Docker setup too.

If you need to check current docs or examples online while building it, go ahead.