Run agent builds and tests on remote Windows VMs
A developer gives coding agents ssh access to WSL instances on remote Windows VMs to run builds and tests instead of on their Mac laptop.
Done withCodex
- What they did
- The developer's employer supplies a Mac laptop plus remote Windows VMs. The agents get ssh access to WSL instances on those VMs and run their builds and tests there. Setup is ad hoc: agents have to pull the latest code, or the developer keeps the main enlistment current on each machine.
- How it went
- Builds no longer run on the Mac, so background agents stop causing video streams to drop during meetings. The developer is happy with it, though the setup is only good enough for now.
- Worth knowing
- Keeping code in sync on the VMs is manual; the developer is considering containers or GitHub Actions to make it cleaner.
Try it yourself with Codex
I'm working in [project folder] on my [Mac/Linux] machine, but builds and tests must run on a remote Windows machine. You can reach it with ssh at [user@host] and use WSL there. Sync the code to [remote path], run [build command] and [test command] over ssh, and report which tests passed and failed with the relevant error output. Do not delete or overwrite anything on the remote machine outside [remote path], and show me any command that would do so before running it.
Want help getting this working? Join the Vibe Coding Academy ↗ Run by Alex Finn, who also runs this site.
Source: HN · Oct 18, 2026
Related use cases
Turn tickets into merged pull requests
Pulls tasks from GitHub Issues or Linear, then runs Claude Code or Codex in isolated Kubernetes pods to carry each one through to a PR.
Handle family errands and recipe sorting with Codex
A communications professional had Codex email her son's doctor for a refill and sort 240 recipes down to 62 family-fit ones into an index, spreadsheet and Drive folder.
Build and publish a BaZi chart reading agent
Built solo with coding agents, an agent computes BaZi charts, interprets them, detects language and takes on-chain payments for readings.
Build a C compiler with a team of parallel agents
Anthropic's engineering write-up on tasking Opus 4.6 agent teams with building a C compiler.
Debug frontend UI bugs with a browser agent
A frontend developer pairs Claude in Chrome with Claude Code to open the app, click through it, read the DOM, check console errors and network requests, and fix UI bugs.
Run sandboxed Claude Code fleet sharing files via MCP
A developer runs a fleet of sandboxed Claude Code instances that share files through MCP tools on a virtual filesystem, with an orchestrator handling AWS storage and audit logs.
Get 5 like this every morning
The best things people got an AI agent to do, each with the prompt to try it.