AI Crowdsourcing Platform
A gamified internal crowdsourcing platform where business teams post AI project ideas and employees pick up tasks, earn recognition, and get credit for contributing.
The Problem
Business teams had ideas for small AI projects but no lightweight way to propose them and find internal contributors — every initiative, however small, needed a formal project team and staffing process to get off the ground.
Why Traditional Systems Failed
Formal project staffing doesn't scale to dozens of small, exploratory AI ideas — most never get resourced because there's no lightweight way to propose them or find willing contributors. Ad hoc requests for help have no tracking, credit, or governance.
Crowdsourcing & Gamification Architecture
Built a role-based crowdsourcing platform where business managers submit ideas and break them into tasks with story-point values; contributors browse open tasks, pick up to three at a time, and move them through an open → in-progress → in-review → completed lifecycle. Completed work earns story points that roll up into tiers and a leaderboard, with anti-gaming caps on point-earning actions and automatic lifecycle transitions (an idea auto-advances to in-progress on its first task pick) to reduce administrative overhead. A four-role access model, with a self-service request-and-approval flow for role upgrades, keeps governance lightweight without a manual provisioning process.
Interactive Architecture
Hover a node for detail · click to isolate its connections
People
Platform
Data
Every contribution and point transaction is logged for governance and tracking.
Key Capabilities
Idea & Task Crowdsourcing
Business managers post ideas and break them into discrete tasks with story-point values for contributors to pick up.
Gamified Recognition
Story points roll up into tiers and a leaderboard, giving contributors visible credit for their work.
Role-Based Governance
A four-role access model with self-service role-upgrade requests keeps provisioning lightweight without manual admin overhead.
Automatic Lifecycle Management
Ideas and tasks auto-transition through their lifecycle (e.g., to in-progress on first pick) without manual status updates.
Full Audit Trail
Every contribution and point transaction is logged for governance and tracking.
What This Enables
- Lightweight sourcing of small AI initiatives without formal project staffing
- Visible recognition and career credit for internal contributors
- Self-service role upgrades instead of manual provisioning
- Governance and audit trail over who contributed what, and when
What This Is Not
- Not a formal project management or staffing system
- Not unstructured crowdsourcing — every task and contribution is tracked and governed
- Not a replacement for dedicated AI/ML engineering teams on larger initiatives
Business Value
Decentralized AI project sourcing and staffing — business teams no longer need a formal project team to scope and launch a small AI initiative, and internal contributors get visible credit and career-relevant recognition for the work.
Key Learnings
- Gamification mechanics (points, tiers, caps on abuse) need explicit anti-gaming rules from day one, or the incentive design gets exploited.
- Automatic lifecycle transitions (auto-advancing an idea's status on first task pick) remove a surprising amount of manual admin overhead at scale.
- Funneling all data access through a single centralized service layer, front and back end, made the codebase far easier to reason about and secure as the role model grew more complex.