GitHub · AI product · Senior · 60 to 90 minutes

Improve Trust in Copilot Code Review

GitHub is a developer platform for hosting code, collaborating on software, and running parts of the software development lifecycle. Answer this product management case and get it graded line by line against a weighted rubric.

Strategic judgment
User segmentation
Metric design
Prioritisation
Executive alignment

The situation

GitHub Copilot code review entered preview in late 2024 and became generally available on GitHub.com Enterprise Cloud in April 2025, and the feature now includes agentic review options. The VP of Product for GitHub Copilot has asked you, a senior PM, for a plan to improve and scale adoption over the next two quarters. Assume the following numbers are illustrative: in paid Enterprise Cloud orgs with Copilot, 30% of eligible repositories have code review enabled, developers react positively to 55% of Copilot review comments, 18% of teams that enabled it later restricted its scope or turned it off, and repos that keep it on merge pull requests slightly faster. Support tickets and customer calls show a pattern, teams like the idea of AI review, but noisy or obvious comments cause developers to ignore or mute the feature, while admins want better controls and proof of value before widening access. Engineering wants broader defaults to grow Copilot usage, while support and trust teams warn that weak comments can train people to dismiss strong ones. You need to recommend how GitHub should tune quality, defaults, admin controls, measurement, and rollout so teams keep Copilot code review on and trust what it says.

What to produce

A 1 to 2 page product strategy memo.

Your answer must

State the core problem as a trust, noise, and retention challenge for a shipped product, not as a first launch or private test plan.
Define the main user and buyer segments you would focus on first, and explain why those segments should get different defaults or controls.
Recommend a 6 month strategy that covers comment quality improvements, repository or language targeting, default settings, and enterprise admin controls.
Define one north star metric, 3 to 5 supporting metrics, and at least 2 guardrail metrics.
Describe an experiment and rollout plan with clear decision points for expanding defaults, pausing, or rolling back.
Identify the main risks and second-order effects, and explain how you would align engineering, applied science, support, sales, and trust leadership.

Constraints

Plan for a 6 month horizon and GitHub.com Enterprise Cloud only, not a separate GitHub Enterprise Server launch.
Assume engineering and applied science capacity for only 2 major product changes and 2 smaller follow on changes in that window.
Do not rely on storing customer code or pull request data beyond current GitHub and Copilot policy commitments.
Go to market budget is limited, so use in product education and existing sales or support channels, not a large custom services program.

Format

Structure your answer as problem, target segments, strategy, metrics, rollout, and risks. Bullets or a simple table are fine if they make tradeoffs clear.