Techniques & Methods
Background Coding Agent
An AI coding agent that works asynchronously in the cloud — assign a task, get a PR hours later.
Also known as: async coding agent,cloud coding agent
A background coding agent is an AI agent that handles software-engineering tasks asynchronously. You assign a task (typically a GitHub issue or written brief), the agent spins up a cloud sandbox with your repo, reads the code, plans the change, edits files, runs tests, iterates on failures, and opens a pull request when done. OpenAI Codex (2026), Devin, Claude Code with background mode, and Cline async are the leading examples. The shift in mental model: from "AI helps me code in my editor" to "I manage AI agents that code while I do something else." Typical power users run 3-5 agents in parallel. Best for well-scoped tasks: refactors, bug fixes, migrations, test backfill, doc updates. Still bad at architecture decisions, open-ended performance work, and tasks needing product judgement.
