Team Culture — Git, Documentation, and Quality Systems
Leading the rollout of Git Flow, architecture docs, pre-commit, and Storybook across a five-person team
Background
A team of five had no development process and no collaboration standards. Code style varied by person, which dragged out review arguments. Git usage differed from person to person, which caused frequent merge conflicts. The absence of system architecture documentation concentrated knowledge in individual developers. New members took a long time to understand the system after joining.
System architecture
- Git and collaboration: adopted a Git Flow strategy, a PR-based review culture, and Jira issue tracking.
- Documentation: a
docs/{architecture, api, components, database}set for the sorting system backend, plus automatic DRF Swagger/OpenAPI documentation. - Code quality: pre-commit on the backend (Black, isort, Ruff), and an ESLint + Storybook + MSW component development environment on the frontend.
How it was built
Establishing the Git workflow. Defined a Git Flow strategy and ran training for the team. Introducing PR-based code review met resistance at first, but sharing the bugs that reviews actually caught changed how the team saw it. Issue and release tracking in Jira was set up alongside it.
The architecture document set. Wrote per-topic documents covering the system overview, Docker services, server structure, components, database, API, and PLC communication. These are split across docs/architecture, docs/components, docs/database, and docs/api. API documentation is generated automatically through DRF Swagger/OpenAPI, so the specification stays current without separate maintenance.
The code quality system. Wired Black (formatting), isort (import ordering), and Ruff (linting) into the backend pre-commit hooks. Style is unified automatically at commit time. The frontend runs ESLint (TypeScript and React Hooks) with Storybook and MSW for component-level development and testing.
Results
- Consistent code style across the team through pre-commit and PR review.
- Knowledge concentration resolved by the
docs/{architecture, api, components, database}set for the sorting system backend plus automatic Swagger documentation. - Shorter UI development and verification cycles through the Storybook and MSW frontend component environment.
- Shorter onboarding for new members and an established collaboration process through Jira issue and release tracking.