The Problem
AI briefs arrived ambiguous — "build something with AI" — with no defined inputs, outputs, constraints, or success metrics, and no clarity on whether a given feature should be a copilot a person drives or an agent that acts on its own.
My Role
Lead Product Manager, on contract, shaping AI product structure across engagements.
The Approach
Defined inputs, outputs, constraints, and success metrics before a line of code was written on each brief. Made the copilot-vs-agent call explicitly for each case, alongside how fallbacks, explainability, and human-in-the-loop handoffs should work. Aligned technical and non-technical stakeholders on what the AI could realistically do, before scope got locked in on the wrong assumption.
The Outcome
- Scope churn cut by aligning stakeholders on realistic AI capability before build started
- Engineering teams unblocked faster, working from specs rather than ambiguous briefs
- A consistent structure for AI product decisions applied across multiple engagements, not reinvented each time