? End of an era for traditional chess engines (AB engines) is coming?

It looks like Stockfish is the last man keeping an honour of traditional chess engines. Leela has been progressing so fast and just after  one year anniversary, she improved from Zero to Hero. She passed almost all level of traditional chess engines including Komodo and Houdini (except Stockfish).

And the current machine learning is still improving. I am curious to see the end of an era of traditional chess engines.

What is AB engine? brute force method engine?

Alpha Beta pruning (advanced serach than minimax) 

https://en.wikipedia.org/wiki/Alpha%E2%80%93beta_pruning

Wow SF had a loss, due to greedily taking free "g pawn" , that allow king side attack. I bet 1400, people dont do such strategical mistake.

I imagine it will be a lot of fun analyzing with a nn engine in the future as it seems they play more rational chess.

And also Leea is weak in tactical position where many branching forcing line need to be calculated precisely. Cos Leela is  x1000 time slower than SF, Leela is more likely to miss tactical holes.

I define tactical-a position is unstable with many branching line

Positional- a position is stable with very narrow branching line

Beyond search horizon, SF will play like 1400(SF built in is 1400), meanwhile Leela will play like 2100 without search (thanks to NN database). But in position where precision and search is extremely important, like endgame , Leela is much much weaker than Stockfish.