What it does
The local stage reduces a 225-cell board to roughly 40 candidates before each Jev call.
Playable demos & bots
Jev gomoku harness
Víťa Vacek's gomoku harness performs tactical filtering locally, then asks Jev to choose among the surviving moves.
Bookmark: Jev gomoku harness
Keep this in your collection.
Leave a noteWhat would you try with this?
: Jev gomoku harness
Only you can see your notes.
Screenshot unavailable. Open the experiment ↗
How you can use it
A developer can borrow this idea for their own board game. The game checks the board itself first. It finds obvious moves like blocks or winning lines. This shrinks the total choices down to a smaller list of good candidates.
Next, the game sends only that short list to Jev. Jev is an outside AI service that picks the final move. This approach uses the AI efficiently. The source only describes this method. It does not provide working software to download.