Honestly, I am not a big fan of steem-engine... It's not a very robust approach, with regards to consensus. It is not guaranteed at all that two clients, executing the same code, will arrive at the same state. Things like floating point operations will lead to problems in consensus and even simply skipping a few blocks will lead to alternative histories of whatever the "smart-contract" is tracking. Wosr than that, there is no way of dealing with consensus problems, except maybe social agreement.

This does not happen in ethereum because smart-contracts are built-in in the consensus layer. This is not to say that this could not be done in steem, but it needs to be done either way more carefully, or at the consensus layer itself. This is in part why SMTs are taking so long to come from SteemIt.