Software Engineering at Google
About this book
Google's account of the practices that let thousands of engineers work in one codebase over decades, organised around culture, processes and tools. The opening argument is that software engineering is programming integrated over time, which is what makes code review, testing, deprecation and dependency management central rather than incidental. Later chapters cover the monorepo, large scale changes, static analysis, continuous integration and the build system. Much of it assumes a scale most organisations will never reach, and the reasoning about why practices break down as time and headcount grow travels well below that scale. Useful to a delivery lead who wants to argue about engineering standards from evidence.
Description via Product Digest.
Topics
Related concepts
- Prompt Regression Testing: Compare Versions Before ReleaseBuild prompt regression tests that compare baseline and candidate versions, expose segment level failures and enforce quality, cost and latency gates.
- Prompt Versioning and Deployment: Canary Releases and RollbackVersion prompts with their model, schema and retrieval settings, then deploy through evaluation, canary monitoring and a tested rollback process.
- Prompt and Context Engineering Practice: A Production PlaybookBuild a repeatable prompt and context engineering workflow with versioned artefacts, evaluation ownership, trace review and a practical capstone project.