Experiment: The hidden costs of waiting on slow build times

This post details an experiment comparing build times and costs across different compute tiers (2-64 cores) using GitHub's larger hosted runners. It quantifies the cost of developer time spent waiting for builds and the additional cost incurred by context switching, concluding that investing in more powerful hardware is generally more cost-effective due to reduced developer idle time and productivity loss. The post includes detailed spreadsheets and visualizations of the cost analysis.