Design for intent
Start from what the person is trying to get done.
- Turn goals into intents an agent can act on
- Decide what the agent infers and what it asks
- Design for vague and changing requests
All courses
Agent Design Engineering
Design agent experiences: the intent, the journey and what good looks like to the people using them.
Next cohortJun 7 – Jul 3, 2027
Length4 weeks
FormatLive cohort
The Laws of SimplicityJohn’s book on design
GitHubDesign engineering
4 weeksLive, with your cohort
1 agent experienceYour capstone, designed and tested
Why this matters now
When software acts on its own, design decides whether people trust it.
Screens told people what they could do. Agents decide and act for them. That changes what designers, PMs and engineers have to get right.
What did the person mean? What should the agent do on its own, and when must it ask? How does someone see what it did, correct it, or undo it?
Agent design is a craft of its own. It sits between interaction design, product judgment and the engineering of a model’s behavior.
This cohort gives you the language, the patterns and the tests to design agents people rely on.
Design the intent, the journey and what good looks like, then test it with real people.
What you'll learn
Everything this certification puts in your hands.
Map the agent journey
Design the whole job, not one reply.
- Map the steps, handoffs and waiting time
- Show progress while the agent works
- Plan where people step in
Earn trust
Make the agent’s work easy to see and to correct.
- Show what the agent did and why
- Design review, undo and approval
- State the agent’s limits in plain words
Define what good looks like
Turn taste into something you can test.
- Write the behavior spec for your agent
- Collect examples of great and poor answers
- Review real transcripts with your team
Prototype with the model
Design in the material itself.
- Prototype agent behavior with AI tools
- Test with real people early
- Change prompts, tools and interface together
Ship and keep improving
Design does not stop at launch.
- Measure task success and satisfaction
- Learn from failures and edge cases
- Keep it simple as features grow
The capstone
Design an agent experience—and prove people trust it.
Pick an agent you are building or want to build. By Demo Day you have its intents, its journey, its behavior spec and a prototype tested with real people.
Your proof: a working prototype, its behavior spec, and what you learned testing it with the people it serves.
1Choose the agent and the people it serves
2Map the intents and the journey
3Write the behavior spec and examples
4Prototype with the model
5Test with real people and refine
6Present it at Demo Day
Curriculum
Week by week.
Dates are estimated and may change.
Week 1Intent and the Agent JourneyJune 7–13+
- Map what people are trying to do
- Design the journey and its handoffs
- Choose your capstone agent
Week 2Trust, Control and CorrectionJune 14–20+
- Show what the agent did and why
- Design review, undo and approval
- State limits in plain words
Week 3What Good Looks LikeJune 21–27+
- Write the behavior spec
- Collect examples and review transcripts
- Prototype with the model
Week 4Testing + Demo DayJune 28–July 3+
- Test with real people
- Refine and measure
- Present at Demo Day
Faculty
Taught by the people doing the work.
John Maeda
Design Engineering at GitHub
John is a design technologist and the author of The Laws of Simplicity.
What's included
Everything that comes with your seat.
Live instruction
Four weeks of interactive teaching with John.
Design reviews
Feedback on your intents, journey and behavior spec.
Capstone feedback
Review of your prototype and what testing showed.
Lifetime access
Return to the content and recordings whenever needed.
Demo Day
Present your agent experience and what you learned.
Certificate
Earn Product Faculty’s Agent Design Engineering certification.
Questions
Is everything taught live?
The Fellowship is built around live, cohort-based learning. You learn directly from the named practitioners, ask questions, complete applied work, and receive the latest insights from people working at the frontier.
Advanced Product Management is included as an on-demand program for strengthening timeless product fundamentals.
How much time will I need each week?
Most cohorts are designed for busy working professionals and require only a few hours per week. The exact commitment varies by program. You'll get the most value by attending live, completing the exercises, and making time for your capstone.
What if I miss a live session?
Attend live whenever possible because the discussions, feedback, and accountability are important parts of the experience.
However, missing an occasional session won't ruin your progress. You can also retake eligible cohorts during your active membership.
Can I retake a cohort?
Yes. You can retake eligible cohorts at no additional cost while your membership remains active. This allows you to experience the updated curriculum, refresh your knowledge, and apply what you learn to a new project as AI evolves.
Can my employer pay for it?
Yes. Many professionals use their learning and development budgets to cover the Fellowship.
We provide a manager-approval email template explaining the practical value to your company. Corporate cohort options are also available for teams. If your company needs an invoice, email support@productfaculty.com and we'll sort it.
Also in the Fellowship
Every other certification, included.
Mar 29, 2027
Nov 30, 2026
Agentic Engineering
Run a fleet of production agents the way you run a team.
Member of Technical Staff, AnthropicHenry Shi
Senior AI Product Manager at GoogleShubham Saboo
AI-Native PM with Claude
Turn Claude and Claude Code into your daily PM team.
Creator of The Product CompassPawel Huryn
New
Feb 22, 2027AI Evals
Turn expert judgment into measurable, repeatable AI quality.
Senior Director, Google ModelsReah Miyara
Most popular
Jan 25, 2027AI Product Management
Ship enterprise-ready AI products: PRDs, RAG, evals, agents.
Product Lead, Codex at OpenAIRohan Varma
Pick your first cohort. Take the rest as they run.
See all 11 courses