Equal workspace

Workspace

Startup routine readyStart setup
Onboarding

Stand up a new organization with the documented startup routine.

The first-run path is now aligned to the source spec: define the organization boundary, assign roles, encode rules, collect preferences, generate assignments, and publish the first schedule.

What success looks like

Managers can complete the first scheduling cycle without outside intervention.

Roles and permissions are clear from day one.

Templates and rules match the real operating environment.

Staff preferences are captured before publishing the first roster.

Startup sequence

First-run checklist

01

Create the organization

Owner

Step ready

Capture the organization name, timezone, and the initial owner account. This becomes the boundary for schedules, users, rules, and analytics.

Organization identity and timezone are defined.

The first privileged user is attached to the workspace.

02

Invite the core team

Owner + Admin

Step ready

Bring in admins and schedulers first so role management, rule configuration, and approval workflows have clear accountability.

Core operators can manage staffing data.

Role boundaries match the documented permission model.

03

Define templates and rules

Admin + Scheduler

Step ready

Set up shift templates, required skills, and the first set of hard and soft constraints before any schedule is generated.

Coverage expectations are encoded as templates.

Legal limits and fairness weights are ready for the solver.

04

Collect staff preferences

Staff + Scheduler

Step ready

Availability, preferred shift types, leave windows, and notes turn a valid roster into a fair one.

Availability is captured before scheduling starts.

Preference data is available for soft-constraint scoring.

05

Generate the first roster

Scheduler

Step ready

Run the optimizer, inspect conflicts and fairness impact, and prepare the first publishable schedule.

Assignments are proposed with conflicts surfaced.

Fairness context is visible before publication.

06

Review and publish

Scheduler

Step ready

Adjust assignments if needed, publish the roster, and notify staff so swaps and follow-up actions can begin.

Published schedules become visible to staff.

The system is ready for swaps, analytics, and audit logging.

Build order

Implementation sequence

01

Database schema and RLS

Create the Postgres tables, indexes, and row-level policies before mutating schedule data.

02

Auth and organization claims

Set up Supabase auth and attach user, organization, and role claims for server-side access control.

03

Core entities and APIs

Implement users, skills, shift templates, schedules, and the supporting server functions.

04

Constraint engine

Build the rule interpreter and assignment optimizer that can propose valid, fairness-aware rosters.

05

Schedule features

Connect the calendar, preferences, fairness ledger, and swap request flow to real data.

06

Realtime, admin, and polish

Add live updates, rule management, analytics depth, accessibility, and final UI refinement.

Roles

Access model

Owner

Full control, billing, organization-level permissions, and escalation authority.

Admin

User, rule, and settings management without organization deletion.

Scheduler

Shift, assignment, swap, and roster generation authority.

Staff

Own schedule visibility, preferences, and swap requests only.

Observer

Read-only access for payroll, auditors, and operational visibility.

Guardrails

Non-negotiables

Legal maximums and rest requirements stay hard constraints.

Approved leave always overrides schedule proposals.

Emergency overrides require audit logging and explicit authority.