There's a confluence of factors that make infrastructure projects such a nightmare in the US, but the big ones seem to be:

-Not institutional knowledge. State DOTs don't retain people who can plan and manage this stuff, it all gets farmed out to contractors or their people get scalped by contractors willing to pay 2-3x the wage the state will pay. So, they're completely at the mercy of contractors.

-Overreliance on contractors and subcontractors. Nuff said. There's a lot of shitty contractors out there whose whole game is to take the taxpayers for as big of a ride as possible, regardless of whether the work actually gets done. Because of Reagan era "reforms" (those are sarcasm quotes, to be clear), we use contractors for all kinds of stuff, and it's easy for shitty contractors to game the system.

-Stations: the US has a hard-on for building large stations, when they're very reliably the most expensive part of building any kind of rail infrastructure. We could substantially reduce rail project costs be re-examining our station designs and opting for more utilitarian choices. I'm not against making stations look nice, mind you, I'm not advocating for a brutalist, khaki concrete cube approach here; just saying that we can make more pragmatic choices than CAHSR's fantasy-future ribcages.