A matching agent that finds a student a study mate for a specific course and time, picks a place both can reach, and emails an invitation once the student approves it. Problem Students who want to study with someone for a course or an exam have to find that person by hand: posting in a class group chat, asking friends, or waiting for a study group to form. The search fails on three things at once: who is taking the same course, who is free at the same time, and where both can get to. Existing tools solve one piece each (course lists, calendars, maps) and leave the student to combine them. Why an agent A workflow cannot do it. The inputs are structured but incomplete. A student's note ("not a library", "somewhere I can talk") needs interpretation, near-tie matches need a judgment call with a stated reason, and the invite has to read like a person wrote it. A human cannot do it at scale. A course coordinator can pair students once at the start of term; a student needs a match for a specific exam next Tuesday. A fully autonomous agent is the wrong answer too. The one irreversible action is emailing another student, so the student approves every message. The agent does the search, the reasoning, and the draft; the student decides.
Built With
- python
- vibe
Log in or sign up for Devpost to join the conversation.