Intelligent Interface Spec for an AI-Assisted Code Review Tool
Overview
What this challenge is about.
Audit current AI-surface inconsistencies and map every AI signal to a confidence band (high / medium / speculative / unverified). Design a unified intelligent-interface spec: surface-selection rules, confidence communication patterns (inline hint vs. side panel vs. modal), and a correction protocol that feeds back into model retraining. Run a Wizard-of-Oz study with 8 senior reviewers reviewing real-world PRs, with a human simulating the AI to test interface variants. Deliver: audit doc, interaction spec, Wizard-of-Oz protocol + recordings, 6-page validation report, and a Figma reference for the front-end team.
The Brief
What you'll do, and what you'll demonstrate.
Restore AI-suggestion acceptance to 38 percent via a unified intelligent-interface spec validated by Wizard-of-Oz testing with 8 reviewers.
Earning criteria — what you'll demonstrate
- Design intelligent interfaces with explicit confidence communication
- Run Wizard-of-Oz studies to validate AI interaction before model integration
- Distinguish trust-building from over-claiming in AI suggestion UX
- Specify human-in-the-loop correction protocols that improve future model quality
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
Careers
Roles this prepares you for.
Real titles. Real skill bridges. Pick the one closest to your trajectory.
Product Manager
Product managers who can design intelligent-interface specs grounded in real trust dynamics own the next wave of AI-feature roadmaps at developer-tools and SaaS companies.
This challenge sharpens
- intelligent-interfaces
- interaction-design
- user-research