Considering the machine was built in the US in 1965, it's not surprising that the machine used imperial units. Based on Wikipedia's table [1], it looks like metrication had not yet begun in a large number of countries by 1965.

Also, if your goal is to build a device that calibrates other machines, and most of the machines in your country (and, likely, most of the machines belonging to the majority of your trading partners) use one system of measurement, then that's probably the system of measurement you build the machine to calibrate against.

[1] https://en.wikipedia.org/wiki/Metrication#Conversion_process

My guess is that when it was constructed 50 years ago, the "device" was engineered with the units that were familiar or already organized into formulas to plug data into. In the early to mid 60's, simple extremely limited dedicated calculators were a new thing and hard to get. Not at all like the tech in our world we take for granted, numerical analysis was much harder to accomplish.

Of course today we have vast computation resources right under our fingertips. How hard is it to translate from one set of units to another? No doubt on issuing a report of their test results, the output will be expressed in whatever units anybody wants, no trouble at all...

The supply chain isn't set up for metric. Steel supply in the US is mostly set up for imperial measurements. If you want 500 metric tons of steel the guys at the foundry will convert it to imperial before the order gets put in. The guys at whatever machine shop is finishing them will convert all the dimensions to imperial before finish machining. The guy doing the final calibration will convert to imperial before doing his job (probably a combination of drilling holes to remove material to get them just under the target weight and then adding precisely weighted chunks of steel for final calibration). Why not just have the conversion be the last step. Saves effort and reduces screw-up potential.