Replicas

What is Replicas?

Replicas is a cloud-based platform that lets you run background coding agents—like Claude Code or Codex—in isolated virtual machines. Instead of keeping your local machine tied up, you spawn a fully equipped dev environment in the cloud, hand off a task, and come back to a pull request ready for review. It's essentially a background worker for your codebase that works with your own subscriptions and API keys.

Who it's for

• Solo developers who want to offload repetitive coding tasks without context-switching or blocking their local workflow.
• Engineering teams that need to parallelize work—assign a refactor, a bug fix, or a changelog update to an agent while the team focuses on higher-priority features.
• Open-source maintainers who manage multiple repositories and could use an automated assistant to triage issues, write patches, and generate PRs from Slack or GitHub.

Key features

Trigger tasks from your existing tools

Replicas integrates directly with Slack, Linear, and GitHub. You can fire off a coding task without leaving your chat or project management tool. The agent picks up the context, works in its own VM, and delivers a PR when done.

Bring your own agents and keys

You're not locked into a proprietary model. Replicas runs Claude Code, Codex, or any coding agent you prefer. You supply your own subscriptions and API keys, so you keep control over costs and model choices.

Sandboxed, real dev environments

Each agent runs in an isolated VM with a complete development environment—code, dependencies, and tooling are all pre-configured. This means agents can actually compile, test, and commit changes without interfering with your local setup.

What stands out

"Delegate to coding agents that work in sandboxed development environments."

The core insight is that coding agents are most useful when they can operate autonomously without blocking your machine or risking your local environment. Replicas makes that practical by giving each agent a disposable, full-stack VM that mirrors a real dev setup. You hand off a task, the agent does the work, and you review the result—no context switching, no resource contention, no cleanup.

Worth checking out if…

You're already using AI coding assistants but wish they could run in the background without hogging your terminal. If you manage a codebase that generates a steady stream of small-to-medium tasks—changelog updates, bug fixes, refactors—Replicas turns those into fire-and-forget operations. It's also a strong fit for teams that want to parallelize development work without adding headcount, especially if you're already living in Slack, Linear, or GitHub.