Upskilling PMs on AI: From Prompt-Curious to Prototype-Shipping

The PM who can prototype has a superpower

Product management has always required translating between user needs and engineering reality. The AI tooling shift changes that translation layer. PMs who can build working prototypes — even rough ones — compress the spec-to-validation loop from weeks to days.

The core toolkit for AI-enabled PMs

• Lovable / v0: UI scaffolding from natural language descriptions. Generate a clickable prototype without touching code.
• Claude (Projects): Maintain context across your entire product domain. Draft PRDs, critique them, iterate on positioning, generate user stories.
• Claude Code: For PMs who want to go further — reading and modifying feature branches, writing test cases, understanding the systems they own.
• Cursor: Pair with an engineer live and contribute meaningfully — even without deep coding fluency.

What changes when PMs can prototype

Three things happen when product teams adopt these tools seriously:

1. Engineering gets better briefs. A working prototype is worth ten pages of specs. Engineers can see the intent, not just read about it.
2. User research gets richer. Testing a clickable flow beats testing a written description every time.
3. Prioritisation sharpens. PMs who've built a rough version of something have a visceral sense of what's hard vs. easy, which changes what they push for.

How we structure PM upskilling programs

Our AI Product track is designed to be delivered as a team cohort, so PMs and engineers run it together. Cross-functional alignment is baked in from day one — both sides leave with a shared vocabulary and shared tools.

The capstone is always a real deliverable: a prototype or a production-quality prompt pipeline that solves an actual problem in the team's current roadmap. No hypothetical exercises.