-
CampusOps: Autonomous Event-Approval & Coordination Agent for Colleges (Team CampusOps)
-
The problem: event approvals mean many stakeholders, rules and deadlines, all coordinated by hand
-
Our solution: an autonomous multi-agent layer that plans, checks, follows up and recovers from failures
-
Key features: natural-language intake, dynamic approval paths, risk detection and autonomous failure recovery
-
Workflow and tech stack: Planner → Compliance → Outreach → Tracker → Recovery, built with React, FastAPI and LLM agents
-
Architecture: a human-in-the-loop system where AI proposes and humans approve
Inspiration
Organizing a college event means chasing approvals from faculty, HOD, principal, finance, the venue team and sponsors. One delayed reply or an unavailable venue can stall everything, and students coordinate it all by hand.
What it does
CampusOps is an autonomous event-approval and coordination agent. A student describes the event in plain language, for example "Tech Fest, 300 students, March 15, ₹50,000". AI agents then plan the approval chain, check college rules, draft requests, track replies and deadlines, and replan when something fails.
Our main innovation is Autonomous Failure Recovery. If the venue is unavailable, CampusOps finds an alternative, checks capacity and timing, and updates the workflow instead of just reporting the problem. Humans approve all official emails and sensitive decisions.
How we built it
This is our idea-stage submission, so the MVP is planned, not built yet. We will build a multi-agent system (Planner, Compliance, Outreach, Tracker and Recovery agents) using an LLM, a Python/FastAPI backend and a React dashboard. The 24-hour MVP will demo one complete workflow using structured college rules, mock venue data, mock approvals and a simulated failure and recovery.
Challenges we ran into
We had to decide how far the agents should act on their own. We chose to keep humans in control of official emails and financial or final decisions. We also had to keep the scope small enough to demo in 24 hours.
Accomplishments that we're proud of
We defined a clear agent workflow, a decision rule for each agent, and a failure-recovery loop that goes beyond ordinary event-management tools. We also designed a dashboard mock-up showing readiness before and after recovery.
What we learned
Real institutional workflows depend on rules, deadlines and people who may not reply. A useful agent has to track state and replan, not just answer questions.
What's next for CampusOps
Support for different colleges, approval hierarchies and event types through configurable rules: college-specific rules today, configurable institutional workflows tomorrow.