Functional DemoAI Platform

AI Role Redundancy Analyzer

An AI-assisted labor capacity simulator that models the workforce needed to hit a production goal.

CapacityPilot new-simulation screen: a grid of industry template cards (Accounting, Agriculture, Call Centers, Construction, etc.) beside a Define-the-goal panel with title, target output, and time period fields. — New simulation: pick industry template and define goal

What it solves

Figuring out how much labor is needed to hit a production goal is hard. Teams have to reconcile current workforce capacity, constraints, and cost against a target, then work out the right mix of hiring, overtime, and productivity gains. That analysis is usually done ad hoc in spreadsheets, without a clear plan, bottleneck view, or cost forecast.

Built for Operations and workforce planners modeling labor capacity against production goals.

Overview

CapacityPilot is a labor capacity simulator. A user picks an industry, sets a production goal, and models their workforce and constraints; the app returns current capacity, the goal gap, an optimized hiring, overtime, and productivity plan, bottlenecks, and a cost forecast. A real-time Capacity Sandbox lets users adjust inputs live, a six-agent AI Operational Council reviews the plan, and industry playbooks compare it to best practice. Results compile into an executive report that can be exported to PDF or shared via a revocable link. Deterministic engines do the math, and the AI is guarded so it never invents numbers.

What this demonstrates

A suite of pure, deterministic, unit-tested calculation engines — capacity, workforce gap, optimizer, bottleneck, labor cost, scenario, roadmap, health, and report — cleanly separated from I/O. An AI council is layered on top with a guard that verifies every figure traces back to engine output, over a production SaaS spine of org-scoped multi-tenancy with five roles, Stripe plan enforcement, and Upstash-backed rate limiting.

What makes it different

The AI never invents numbers — a guard verifies that every figure in generated output traces back to a deterministic engine result.

Pure, deterministic, fully-tested calculation engines
Core math never lives in UI components
AI is an advisor layered on verified numbers, not their source
Boots and degrades gracefully without AI, Stripe, or Redis keys
Organization-scoped with five roles
Separation of concerns across engines, services, validation, and UI

Core capabilities · 15

Capacity Engine
Computes current production capacity from the workforce model.
Workforce Gap Engine
Measures the gap between capacity and the production goal.
Optimizer EngineHiring / Overtime / Productivity
Plans hiring, overtime, and productivity to close the gap.
Bottleneck Engine
Identifies the constraints limiting output.
Labor Cost Engine
Forecasts the cost of the plan.
Scenario Engine
Compares alternative capacity scenarios.
Health Engine
Scores operational health.
Roadmap Engine
Sequences the plan over time.
Capacity Sandbox
Adjust inputs and see capacity recompute live.
AI Operational Council6-agent AI Council
A six-agent Claude review of the plan.
Industry Playbooks
Compares the plan against industry best practice.
Industry Templates
40 templates with per-industry operational inputs.
Executive Report
Compiles results into an exportable, shareable report.
Billing & Plan Limits
Stripe subscriptions with enforced per-plan quotas.
Rate Limiting
Upstash Redis limiter with an in-memory fallback.

How it works

Sign in and start a new simulation
Pick an industry and answer industry-specific operational questions
Review capacity, gap, plan, bottlenecks, cost, scenarios, and health
Adjust inputs live in the Capacity Sandbox
Run the AI Operational Council and compare to an industry playbook
Generate an executive report and export to PDF or share a revocable link

How the model works

Capacity & Goal Gap

Current capacity
Production capacity derived from the workforce model.
Goal gap
The shortfall between capacity and the production goal.
Optimized plan
The hiring, overtime, and productivity mix to close the gap.
Bottlenecks
Constraints that cap output.
Labor cost forecast
Projected cost of the plan.
Operational health
A score of the plan's operational health.

Key screens

New Simulation
Simulation Results
Capacity Sandbox
AI Council
Industry Playbook
Executive Report
Settings & Billing

Data model

Organization
User
Session
IndustryTemplate
Simulation
Goal
Team
Role
Constraint
Scenario
OptimizerResult
Bottleneck
Report
Comment
BillingPlan

Screens · 2

Agriculture plan simulation with optimization-mode tabs, Goal/Current Capacity/Recommended Plan cards showing 1,000 vs 2,100 acres surplus, and a Capacity Sandbox with what-if levers and projected output, cost, and utilization tiles. — Simulation detail with capacity sandbox levers

Lower half of a simulation: goal and constraints forms on the left, AI Recommendation Center showing capacity health 89, 95% goal probability, and bottlenecks, above a scenario-comparison table of staffing options with cost per unit. — AI Council recommendations and scenario comparison

How it is built

A Next.js 16 App Router application in TypeScript, with business logic in server-only services that orchestrate the engines and a Prisma database on PostgreSQL. Anthropic's Claude powers the AI council behind a provider interface with a deterministic fallback, Stripe handles subscriptions, and Upstash Redis provides rate limiting.

APIs and services

Anthropic
Stripe
Upstash

Classification

Capabilities: AI Automation Financial Modeling Analytics Reporting Dashboards Authentication Subscription Billing Multi-Tenant Architecture Role-Separated Interfaces
Technologies: Next.js React TypeScript Node.js Prisma PostgreSQL Tailwind CSS Zod Docker

Building something with similar AI Automation, Financial Modeling, Analytics functionality?

Start a project and reference this one — the conversation begins with what you have already seen.

Start a project