What's interesting about this is the conscious decision to throw money at the government IT problem. Meaning, services like AWS Lambda get very, very expensive at scale. Nationwide government-scale is probably pretty high scale, and I shudder to think what the AWS bill would be.

Yet, most government IT projects run very high costs to begin with anyway. Indeed, for all we know, they probably saved money, even with the high costs of running this at scale on Lambda. And it'll be worth it if the service is actually functional.

Just interesting to think about the cost of introducing stability as a standard before eventually the pendulum will swing back and get people to think about cost optimization.