Most good data stories start with a interesting question. If the average request latency went down by a further 100ms, by how much could we expect user engagement to increase? How can we detect evidence of corruption of government officials given a list of all bids nationwide for the building of new roads and repair of existing ones? Can we identify a new pandemic in the making given a timeline of common search terms? Often though, we know we have the data, but we don't even know what questions the data might help answer, or how the story will unfold. From the data scientist who scratched an itch on an idle afternoon, to a low latency, high availability, real-time analysis deployed on hundreds of machines, the story typically involves lots of rewrites, meanderings and building out of a lot of code and utilities to improve the precision, speed or scale of the analysis.