What I suspect is this 17% is the exact sub-set it needed to hack together to make the goal (running some example website) a reality as this is what those dodgy weasels do if you let them. Then you get to spend 200x the time to fill in the rest of the "speculative features deferred due to no real consumer" on top of whatever dodgy system they made up, which is usually whatever is easiest/closest to the literature instead of the actual intended design. Lots and lots of fun to be had doing the full-pipeline refactors to add that last 2% which need support from tip to tail.
The article doesn't go into how they managed the AI context when implementing things but I would not be surprised if it was done in a methodical way, 80% - 90% of the test could have passed.
It's all in good fun, though... probably?