PM Control Room

Inspiration

Project managers rarely lack data. The real problem is that project evidence is scattered across plans, risks, actions, decisions, financials, reports, and stakeholder updates. By the time those signals are consolidated, the project may already be in trouble.

PM Control Room was inspired by a simple question: What if a project manager could see what matters, understand why it matters, and decide what to do next from one intelligent workspace?

The goal was not to replace the project manager. It was to give them better intelligence while keeping human judgment and authority at the center.

What it does

PM Control Room is an AI-powered project intelligence workspace for waterfall and hybrid delivery.

It brings project evidence together and follows a controlled decision workflow:

Detect → Explain → Recommend → Approve → Act

The application helps users:

• Monitor multiple projects from a central dashboard
• Identify emerging risks, schedule variance, and delivery concerns
• Understand why a project requires attention
• Receive explainable, evidence-based recommendations
• Review and approve actions before they are executed
• Navigate project plans, risks, actions, decisions, reports, procurement, and work items
• Support both waterfall and hybrid project delivery
• Maintain the project manager as the final decision authority

Rather than presenting another collection of disconnected metrics, PM Control Room turns project evidence into understandable and actionable intelligence.

How I built it

PM Control Room was designed and built with Codex and GPT-5.6 during OpenAI Build Week.

Codex supported the full development lifecycle, including:

• Translating the product concept into a working application
• Designing the information architecture and user experience
• Building the project dashboard and supporting modules
• Implementing project intelligence and recommendation workflows
• Creating realistic project scenarios and demonstration data
• Testing, reviewing, and refining the application
• Preparing the repository, documentation, and deployment

The application was built as a full-stack Codex Site. Its architecture uses Cloudflare-compatible services, including D1 for structured application data and R2 for document storage.

GPT-5.6 was especially valuable for reasoning across project evidence, explaining delivery signals, and generating recommendations that remain subject to human review and approval.

Challenges

The greatest challenge was converting real project-management complexity into an experience that remained understandable and useful.

Project health cannot be represented responsibly by a single score or color. Schedule, risk, procurement, dependencies, decisions, and stakeholder impact must be considered together. The application therefore needed to explain its reasoning instead of simply labeling a project as healthy or unhealthy.

Another challenge was balancing AI assistance with governance. PM Control Room needed to provide meaningful recommendations without silently taking control away from the project manager.

Building and refining a complete working prototype within the Build Week timeline also required rapid iteration across functionality, data, interface design, testing, and deployment.

Accomplishments

I am proud that PM Control Room became more than a dashboard. It demonstrates a practical model for responsible AI-assisted project delivery.

The final prototype:

• Converts project evidence into explainable recommendations
• Keeps human approval embedded in the action workflow
• Supports waterfall and hybrid delivery approaches
• Presents complex project information through a polished control-room experience
• Demonstrates how Codex and GPT-5.6 can help a nontraditional software builder turn professional experience into a functioning product

What I learned

This project demonstrated how far a domain expert can go when paired with an advanced coding agent.

Codex made it possible to move rapidly between product thinking, implementation, testing, and refinement. At the same time, the experience reinforced the importance of clear requirements, frequent validation, and human oversight—both when managing projects and when building with AI.

The strongest results came from treating Codex as a capable development partner while keeping product decisions, acceptance criteria, and final authority with the human builder.

What's next

The next phase is to expand PM Control Room from a contest-ready prototype into a production-ready project operating system.

Future development could include:

• Configurable organizational governance and delivery standards
• Integrations with common project-management and collaboration platforms
• Portfolio-level intelligence and cross-project dependency analysis
• More advanced reporting and executive brief generation
• Custom AI policies and approval thresholds
• Expanded document intelligence and evidence traceability
• Role-based access and enterprise security controls

The long-term vision is straightforward:

Project intelligence. Human authority.

Built With

• chatgptsites
• cloudflared1
• cloudflareworkers
• codex
• gpt-5.6
• react
• typescript
• vinext
• workers