Each role does one job, and only that job.
A Portfolio Strategist reads the portfolio and proposes where to invest. A Product Planner turns an accepted direction into product intent, a Blueprint Architect turns that intent into a Blueprint, and a Reuse Architect checks what the portfolio can already satisfy before anything new is created. Below them sit the Critic, the Engineering Manager, the coding agents, the Code Reviewer, QA, Security and the Release Manager. Each one has its own inputs, its own scope and a list of actions it must escalate rather than take.
Strategy roles never touch code
The Portfolio Strategist and Product Planner work on intent: which products matter, what a product should do, what the business gets back. What they produce is a case, with options and consequences, put to a person who accepts or declines it. Nothing from that layer reaches a repository until it has become a Blueprint, been composed from capabilities and cleared policy.
Architecture roles describe meaning, not stack
A Blueprint Architect sets out entities, workflows, policies and capabilities — what the software means. How that becomes a running system is left to the generators and target adapters, so business truth survives a change of framework, database or cloud. Alongside it, the Reuse Architect asks the prior question: can the portfolio already do this, and is a new capability genuinely needed?
Delivery roles end at the release gate
Engineering Manager, coding agents, Code Reviewer, QA and Security carry work from Blueprint to pull request to verified build. The Release Manager assembles what the last gate needs — validation results, tests, review outcomes, approvals on record — and the release goes out when policy is satisfied and the required human signature is there. No role in that chain reviews or approves the step it performed itself.
In this capability
- Written scope per role
- Architecture and reuse roles
- Delivery, QA and release roles
- Role recorded on every decision
AI Workforce
The rest of AI Workforce
Separation of Duties
An agent never approves its own output. Proposal and review are separate roles, held by separate agents, every time.
Model Routing
Choose which model does which job. Two rules hold: no model reviews its own work, and critical workflows use more than one.
Agent Permissions
An agent holds a scope, not a key to everything: what it may propose, what it may change, what it must escalate.
Decisions, Not Transcripts
The interface shows the decision, its reasoning, its impact and the options — not a scroll of agent conversation.
Stop managing prompts. Start managing software strategy.
Connect your repositories. Let Foundariq understand what you have already built. Then tell it what your portfolio should build next.
Read-only to start · Your code stays in your GitHub · [email protected]