Writing C++ with Jev by picking valid next tokens
The author experimented with TypeSafe's Jev decision/classifier model for C++: at each step it was given only grammatically valid next tokens to pick from, avoiding free-form generation.
A developer shares deploying Jev code to production, used in several places within a new feature, including checking if a lesson warrants a practice sandbox.
Just deployed some Jev code to prod for a real use-case. It's used in quite a few places within a new feature I'm rolling out. The least interesting piece is checking to see if a lesson is worthy of a practice sandbox: