Personal case study on using a network of specialised sub-agents with RovoDev CLI to improve reliability on a growing codebase. Covers why a single ‘do-everything’ agent started dropping tests and quality checks, how an orchestrator-plus-sub-agents pattern (backend, frontend, tests, performance, architecture, docs, Git/quality) was set up, and the impact on test coverage, quality gates, and ‘fire-and-forget’ development workflows.

Atlassian provides insights into software development collaboration, project management, and team productivity, offering documentation, tutorials, and best practices for using Atlassian's suite of tools, including Jira, Confluence, and Bitbucket. By exploring Atlassian's curated content, developers can learn about agile methodologies, version control workflows, and continuous integration practices for managing software development projects effectively. Whether you're working in a small startup or a large enterprise, Atlassian offers resources to streamline your collaboration and drive project success.

Atlassian

A developer shares their experience using RovoDev CLI on a side project, where a single AI agent initially worked well but became unreliable as the codebase grew, dropping test coverage to 20% and missing quality checks. They solved this by implementing a network of specialized sub-agents (backend, frontend, testing, performance, architecture, documentation, Git/quality) coordinated by a main orchestrator agent. This pattern improved reliability, maintained test coverage, enforced quality gates, and created a more predictable "fire-and-forget" development workflow. The approach allows using stronger models for complex decision-making agents and lighter models for simpler tasks.

How Specialised Sub-Agents Enhanced RovoDev CLI Reliability for My Project

When the “One Big Agent” Started to Break Down

A Different Approach: Specialised Sub-Agents

What Working With This Setup Actually Feels Like

Why This Has Been a Game Changer (For Me)