Overall, I think they could have reduced engineering staff in the long term by adopting mainstream open source technologies instead of their own custom (albeit also oss) database/filesystem/streaming system/rpc library/spark stack.
But those migrations would have taken years. Instead, slashing staff that fast is likely to cause slow rot of their ad network, and overall system resiliency.
Can you share the graph of Twitter reliability over years that you plotted to arrive at this conclusion? Thanks.
Yet somehow GitHub's outages are tolerated with something breaking over there every week.
I have not seen anything else break more times than GitHub even having Microsoft footing the bill and they should have a much higher uptime than Twitter / X which eliminates any excuses over such frequent outages.
But it turns out that GitHub's reliability is far worse than Twitter / X.
This isn't exactly a hard concept to grasp though: it's entirely possible to use GitHub and still be productive when it's temporarily offline; Git doesn't stop working on your local machine because GitHub is down.
It's literally impossible to use Twitter when it's temporarily offline. There is no "offline" mode for Twitter.
There is also the argument to be made that (a) what GitHub does is orders of magnitude more complex than what Twitter does, and (b) I don't recall anyone buying a controlling stake in GitHub and then loudly declaring that it doesn't need all those staff and firing (IIRC) all but one of the Ops team.
Twitter needs to be serving ads to make money (primarily).