I've been vaguely following OpenCog (https://wiki.opencog.org/w/The_Open_Cognition_Project) for years and I know the creator Ben Goertzel is a mathematical prodigy with some interesting ideas. I've read some of his book, Engineering General Intelligence, and I have an abstract understanding of OpenCog's mind-agents and cognitive synergy.But I'm not familiar enough with the state of AI to really know if the OpenCog architecture is sound. Has anyone created anything interesting with OpenCog? Does the hypergraph architecture seem less relevant in the era of deep learning? Can someone smarter than me offer their thoughts?