dev-alex) in Settings, Environments and gives it a monthly budget. The new member then connects their agent and picks that environment at sign-in; the agent works only there.
”I just joined the team. What resources and templates are available?”
The agent gives an overview:list_templates: shows available instance templateslist_environments: the environment this agent works inlist_gpu_types: shows GPU options with pricingbilling_balance: checks workspace credits
Welcome to the workspace. Here’s what’s available: Instance templates:Your environment: dev-alex (picked when you connected me) GPU options (live hourly rates come fromlist_gpu_types):Workspace balance: $1,247.80 For development, I’d recommend starting with an RTX 4090 using the ubuntu-devbox template.
”Deploy my first instance.”
The agent sets up the new team member:create_instance: deploysalex-devboxwith RTX 4090 using the devbox templateinstance_status: waits until runningget_instance: retrieves connection details
Connect:
- Deployed alex-devbox (RTX 4090, 24 GB VRAM)
ssh root@203.0.113.55Pre-installed: Python 3.11, CUDA 12.4, pip, git, htop, tmux. Your instance runs in dev-alex, separate from staging and production. All environments share the workspace credit balance. Terminate when done to stop billing.
”What is in my environment so far?”
The agent lists what it can see, which is its own environment:list_instances: the instances in dev-alexlist_volumes: the volumes in dev-alexlist_ssh_keys: the workspace’s SSH keys
dev-alex (your environment)Team SSH keys: 4 keys registered (macbook-pro, ci-deploy, workstation, alex-laptop) I only see dev-alex. Staging and production are in the dashboard; to have me work in one of them, reconnect Runcrate and pick it at sign-in.
- alex-devbox (RTX 4090): running