Product design question
Design an enterprise agent marketplace on the Gemini Enterprise platform.
- Google DeepMind
- Product design
- Hard
Practice this question out loud. An AI interviewer asks it, follows up like a real interviewer would, and scores your answer. Type or speak.
Start a mock interview on this question · Mock interview from a job description
What this question tests
Platform product design: building a marketplace model for enterprise AI agents, balancing discoverability, trust, and governance.
How to approach it
- Clarify the goal: let third-party developers and internal teams publish reusable agents that enterprise customers can discover and deploy within Gemini Enterprise.
- Design discovery: a categorized, searchable catalog with clear descriptions of what each agent does, what data it touches, and reviews from other enterprise users.
- Design trust and governance: a review and certification process before an agent is listed, plus admin controls letting IT restrict which agents employees can install.
- Design monetization: a revenue-share model for third-party developers, similar to established app marketplaces, to incentivize a healthy supply of agents.
- Add usage transparency: enterprise admins can see which agents are installed, what permissions they hold, and usage volume, for security and cost oversight.
- Define success as number of actively used third-party agents per enterprise account, and marketplace revenue, balanced against zero security incidents from listed agents.
What a strong answer includes
- Treats trust and governance as first-class design requirements, not an afterthought, appropriate for an enterprise audience.
- Proposes a concrete certification and review process before agents are listed, addressing the core enterprise concern about third-party code.
- Names a specific monetization model (revenue share) that is proven in other platform marketplaces.
- Gives IT admins visibility and control, which is a hard requirement for enterprise platform adoption.
- Balances growth metrics (active agents, revenue) against a security guardrail metric.
Common mistakes
- Designing an open marketplace with no review or governance process, unrealistic for an enterprise buyer.
- Ignoring admin visibility and control needs.
Likely follow-up questions
- How would you vet a third-party agent before listing it?
- How would you handle an agent that later turns out to be unsafe?
- How would you price the revenue share to attract quality developers?
More product design questions
- How would you improve Jules (coding agent) to compete with Cursor and Claude Code?Google DeepMind · Product design · Medium
- How would you design Project Mariner (web-browsing agent) to safely book things on a user's behalf?Google DeepMind · Product design · Hard
- Design a consumer feature using Project Astra (real-time multimodal assistant).Google DeepMind · Product design · Hard
- How would you improve Gemini's integration across Google Workspace (Docs, Gmail, Sheets)?Google DeepMind · Product design · Medium
- Design a TV for a car.Google · Product design · Hard
- Design a refrigerator for the blind.Google · Product design · Hard
More questions from Google DeepMind
Learn the skill behind it
Chapters of the AI PM course that teach what this question tests.
- Chapter 4: Discovery and strategy for AI products
- Chapter 7: AI UX and human oversight: design for a system that is wrong sometimes
- Chapter 14: Get the job: the AI PM interview loop