@github: AI coding agents can generate features fast, but can package them in one giant, hard-to-review pull request. Next time,…

AI coding agents can generate features fast, but can package them in one giant, hard-to-review pull request.

Next time, try breaking the work into a stack of focused, ordered PRs.

With stacked pull requests, each layer tackles one concern, like data, API, application wiring, or UI, and builds on the layer below it.

That means smaller diffs, clearer context, and more targeted reviews.

Coding agents can help build and manage the stack, while CI evaluates every layer against the stack’s base.

Devs can create, submit, rebase, and sync the stack using the gh stack CLI.

For reviewers, the workflow is simple:

Read top-down to understand the end goal Review bottom-up to validate each dependency in order

Instead of one 1,700+ line review, you get a series of focused checkpoints.

Learn how to turn one giant AI-generated pull request into a clean, reviewable stack and keep every layer in sync as feedback lands.

@GoogleAI’s Gemini 3.7 Flash is now generally available and rolling out in GitHub Copilot.

Early testing shows: improvements in web and app development and agentic coding workflows improvements in code quality, final-output presentation, codebase research, and verification during complex coding tasks

Try it out in the GitHub Copilot app, CLI, and @code. https://github.blog/changelog/2026-08-13-gemini-3-7-flash-is-now-available-in-github-copilot/…