Study group mode lets up to four students share a session. Each student attempts the same problem independently. Each student's Explain-back is visible to the group. Raw attempts are private.
Why this design: copying another student's raw attempt is the same failure mode as copying an AI's answer. But comparing Explain-backs — reading how a peer reasoned about the same concept — is one of the highest-leverage study activities that exists.
Rules of thumb: one problem per meeting, minimum 15 minutes of independent work before anyone shares, then take turns reading each other's Explain-backs aloud.
The goal is not to be right first. The goal is to identify why your reasoning differed from the peer whose Explain-back you found most useful. That's the study group's real product.