Running a large course cohort in Discord requires more than creating channels and roles. Students need to be placed into the right teams, each team needs its own workspace, and team assignments must remain accurate as new members join or workflows are re-run.
This use case explores how a connected automation can create Discord team infrastructure, maintain a live record of members and roles, assign eligible students to available teams, apply permissions, and provide recovery tools for managing exceptions.
The Problem
Running a large no-code or low-code cohort in Discord can involve hundreds of students who all need to be placed into exactly one team across dozens of channels and roles.
Without a structured process, course administrators may need to manage assignments through spreadsheets, manually update Discord roles, monitor capacity, and resolve duplicate assignments or missing permissions. Re-running an assignment process can also create risks such as duplicate roles, duplicate channels, or students being assigned to more than one team.
The Solution
The Discord Course Team Manager is a community provisioning and team-assignment system built for Divverse’s own course cohort.
It creates team infrastructure in controlled batches, syncs Discord member and role data into Postgres, rebuilds an eligible pool of unassigned students on every run, assigns students to available team slots, and applies the assignments to Discord through a separate scheduled workflow.
The system also includes role-to-channel permission management, onboarding support, debugging queries, and recovery tools for failed or incomplete assignments.
How the Agent Works
- Create team infrastructure. The workflow creates Discord team channels and matching roles in controlled batches to avoid large, difficult-to-manage setup runs.
- Sync Discord data. Member and role data are regularly synced from Discord into a central database so assignment decisions are based on the latest available information.
- Build an eligible student pool. The workflow identifies students who are eligible for assignment and excludes anyone who is already assigned, pending assignment, or part of an excluded system or administration role.
- Assign students based on available capacity. Eligible students are distributed into open team slots while the workflow tracks team capacity and assignment status.
- Apply assignments and permissions. A separate workflow updates Discord roles after the database assignment is confirmed, then grants each team role access to its corresponding channel.
- Monitor and recover from issues. Built-in checks identify duplicate assignments, overfilled teams, incomplete Discord updates, and students who need to be returned to the eligible pool.
Technical Workflow
1. Team infrastructure setup, channels
A code node generates 300 team names and splits them into six batches of 50. The workflow is run six times, each time selecting a different batch and Discord category (Alpha through Zeta), so channels land in the right section without one giant unmanageable run.
2. Team infrastructure setup, roles
A mirrored pattern generates matching role names in batches (covering ranges like 1–100, 101–150, 151–200), then creates them via the Discord API in throttled batches of 5 with a 10-second wait between batches.
A sticky note explicitly warns that rerunning this step will create duplicate roles with the same name, so it is meant to be run once per batch.
3. Cleanup tooling
Separate Delete Roles and Delete Channels branches pull every role or channel from the guild, filter down to a hardcoded set of team names, and delete the matches in the same throttled batch pattern. This is used to tear down and cleanly rebuild test batches during setup.
4. Member and role sync (hourly)
A scheduled workflow paginates through the entire guild member list (1,000 per page via a cursor) and the full role list, upserting both into Postgres members and roles_status tables keyed by Discord ID, so the database always reflects live Discord state before any assignment logic runs.
5. Eligible pool construction
A single query truncates and rebuilds an eligible_pool table on every run. It excludes anyone who already has a pending or completed team assignment, and anyone holding a system or admin role (Admin, Moderator, Mentor, MEE6, Member, bots).
This table is the core anti-duplication guard. Nobody in it can already be on a team.
6. Capacity-aware role assignment
For each team role that is not yet filled, the workflow computes remaining slots (capacity minus already-assigned), randomly selects exactly that many students from the eligible pool, inserts them into role_assignments as pending, immediately removes them from the eligible pool, and updates the role’s fill status.
The workflow loops role by role until every open team is filled or the pool runs out.
7. Delayed Discord write-back (separate scheduled workflow)
On its own hourly schedule, offset five minutes after the assignment workflow, a second workflow reads every pending row from role_assignments, calls Discord’s member-role endpoint in batches of 3 with a 25-second wait between batches, and flips each row to completed only after the Discord call succeeds.
This deliberately separates “decided in the database” from “actually applied in Discord” so a mid-run failure can never leave the two out of sync.
8. Role-to-channel permission sync
A separate branch fetches all channels and all roles named with a matching team- prefix, pairs each channel to its identically named role, and grants that role view and send permissions on its own channel.
This keeps access scoped so each team only sees its own space.
9. Baseline “Member” role rollout
A parked branch paginates all guild members and grants everyone without an excluded role a default Member role. This is used for onboarding students who join before any team-assignment logic has run for them.
10. Debugging and recovery tooling
A dedicated set of maintenance queries covers the failure modes that matter most: counting duplicate assignments, flagging roles that exceeded their 15-person capacity, finding members who look “completed” but were never actually placed on a team, and a recovery query that resets anyone whose Discord write silently failed back into the eligible pool so the next run picks them up again.
The workflow also includes ad hoc username lookups for one-off troubleshooting.
Technology and Integrations
Built with: n8n, Discord Bot API, Postgres, scheduled triggers, and five loosely coupled workflows for infrastructure setup, cleanup, member and role sync, assignment, Discord write-back, and permissions.
Outcome
The Discord Course Team Manager creates a more controlled way to manage team assignments across large course communities.
Students can be distributed across team channels automatically, while the database remains the source of truth for who has been assigned and where. Separating assignment decisions from Discord write-back helps prevent duplicate assignments and reduces the risk of Discord API failures leaving records out of sync.
The built-in debugging and recovery tools also make it easier for course administrators to identify and resolve assignment issues without manually auditing the entire Discord community.
Explore Custom AI Automation for Your Business
The Discord Course Team Manager is one example of how AI automation and connected workflows can support large communities, training programmes, onboarding processes, and internal operations.
Divverse Labs designs and builds custom AI agents, automation workflows, internal tools, and connected systems for a wide range of business processes.
From sales, marketing, recruitment, customer support, and reporting to finance, operations, training, and internal team workflows, each solution is designed around the way your business works.
Request a Quote for a Custom AI Automation Solution to get started.