When I do annual reviews at my company we have someone be an unbiasing champion. They are looking for things like recency bias, familiarity bias, affinity bias and other cases where a speaker might be appealing to something that isn't demonstrated by data. That person's job is to call people out and say, "you just said you really like that engineers personality, but can you frame that against one of the annual review criteria?" Remove the gut check, focus on the days

You might say that we therefore looking at reviews through an unbiasing lens.

Using an antiracist lens could mean they had one person in the room whose responsibility was to tie the conversation back to data about employee performance rather than, "well, I have a hard time understanding person x, let's fire them."

It's not just black and white, there are many ways this could be an innocent policy by Twilio. It could be what you describe, it could just be "we nominated someone to speak up in the meeting, but didn't really listen to them much".

My point is that we don't know. You've made up your mind, but it's not that simple.