RecNet is a human-driven recommendation system for academic readings. RecNet implements a mechanism similar to contemporary social networks, but it is designed to be impoverished in certain ways through information bottlenecks that increase communication cost. This is intended to limit the amount of time the system consumes from its users, while increasing the quality of information passed. The RecNet mechanism was initially outlined by Yoav Artzi in a Substack post. RecNet is currently in initial development stages.

I joined the team behind RecNet at January 2024, work as full-stack developer and designer.

Github