How to run remote coding agents
Blog post from Northflank
Remote coding agents run on infrastructure separate from a developer’s laptop, allowing them to continue long tasks such as refactors, migrations, testing, and dependency upgrades after the user disconnects. They can be hosted on self-managed servers or VMs accessed through SSH, where developers must provision, secure, maintain, and monitor the environment, install agent tools, manage credentials, and use utilities such as tmux to preserve sessions. The text contrasts this approach with Northflank Cloud Harnesses, managed isolated cloud workspaces supporting agents including Claude Code, Codex, Cursor, OpenCode, Pi, and custom CLI-based agents, with persistent storage, browser and SSH terminal access, configurable resources, collaboration features, and per-second billing that stops when environments are paused. Each Harness uses microVM isolation and can run in Northflank’s cloud or within customers’ AWS, GCP, Azure, Oracle, or CoreWeave accounts through BYOC, while the text emphasizes limiting credentials, using separate branches, and reviewing agent-generated changes in pull requests before merging.
| Trend | Post Mentions | Total Month Mentions | Posts | Companies | MoM |
|---|---|---|---|---|---|
| Secrets Management | 1 | 3,093 | 620 | 154 | +51% |
Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.