Design and validate a product experience with AI
From user problem to a prototype tested with real users, then to an outcome-based roadmap, using AI to move faster without replacing users.
Course designed by Alex Dionisio, Senior Product Manager. Updated September 28, 2026.
- 77 steps
- ~2 hr 30 min
Level: Junior · Some basics are useful: introductory knowledge of the topic.
Go from a user problem to a prototype tested with real users, then to an outcome-based roadmap and a spec your development team can build from. You use AI to prepare, synthesize and prototype faster, without handing it what requires evidence: what real users say and do.
What you will be able to do
- Write a user-centered, testable problem statement backed by discovery evidence.
- Prepare, run and synthesize unbiased discovery interviews, using AI for preparation and synthesis without delegating judgment to it.
- Build a persona and job stories grounded in real quotes, then choose a target opportunity in an opportunity solution tree.
- Map a flow with its edge cases and turn it into low-fidelity wireframes that cover every screen state.
- Choose a prototype's fidelity and tool (including an AI prototyping tool), run a usability test and prioritize the problems observed.
- Prioritize a backlog with an explicit method (RICE, MoSCoW, value / effort), making the score's assumptions visible, and justify the trade-off.
- Build an outcome-based roadmap with success criteria, and write a handoff spec with acceptance criteria.
Prerequisites
- Having taken part in designing or launching a digital product, or leading such a project
- Regular use of a consumer AI assistant (ChatGPT, Claude…)
- Recommended, not required: Learn to Prompt (discovery and synthesis prompts).
Syllabus
What will I produce in this course, and in what order?
- Course overview3 steps
Objective · By the end of this overview, you will know what you are going to produce (an experience dossier made of 11 deliverables) and why the four modules follow the order understand, design, validate, then plan.
Which problem, for whom, and what evidence do we have?
Objective · By the end of this lesson, you will be able to write a user-centered problem statement with no hidden solution, and state the evidence that supports it or would contradict it.
Objective · By the end of this lesson, you will be able to write an interview guide based on past behavior, run an interview without leading the answers, and use AI to prepare and synthesize while checking what it produces.
Objective · By the end of this lesson, you will be able to build a persona from your interview synthesis and write job stories that describe the situation, the motivation and the expected outcome.
Objective · By the end of this lesson, you will be able to link a measurable product outcome to opportunities and several solutions in an opportunity solution tree, choose a target opportunity and identify the riskiest assumption to test.
What does the experience look like, including when things go wrong?
Objective · By the end of this lesson, you will be able to map the happy path of a job story, then its edge cases (error, cancellation, missing data, permissions), as a user flow the whole team can read.
Objective · By the end of this lesson, you will be able to produce low-fidelity wireframes that cover every state of a screen, and critique them with Nielsen's heuristics before showing them to users.
Does it work for real users?
- Prototype, including with AI6 steps
Objective · By the end of this lesson, you will be able to choose a prototype's fidelity and tool according to the question being tested, write the brief for an AI-generated prototype and state what that prototype does not prove.
Objective · By the end of this lesson, you will be able to write a usability test script with non-leading tasks, run a think-aloud session and rank the observed problems by severity to decide what to fix before retesting.
In what order to ship, how to communicate it, and how to hand off without losing anything?
Objective · By the end of this lesson, you will be able to score ideas with RICE while making the assumptions behind each factor visible, frame a timebox with MoSCoW, and justify a trade-off that departs from the score.
Objective · By the end of this lesson, you will be able to build a Now / Next / Later roadmap organized around problems and desired outcomes, with an MVP defined by its hypothesis and a success criterion per horizon, and to defend it against requests for dates.
Objective · By the end of this lesson, you will be able to write a one-page spec (intent, constraints, flow, states, acceptance criteria) and assemble your experience dossier to hand it to the development team.
Glossary terms
Public definitions of the key concepts in this course:
Verified sources
The fast-moving facts of this course (tools, models, regulation) are checked against these sources, then reviewed according to how quickly they go stale.
- Adobe XD (Wikipedia), product status and cited sources · verified September 28, 2026
- Anthropic, Introducing Claude Design by Anthropic Labs (April 2026) · verified September 28, 2026
- Claude Academy, Prototype AI-powered apps with Claude artifacts · verified September 28, 2026
- Figma Make · verified September 28, 2026
- v0 (Vercel) · verified September 28, 2026