Development Environments

This guide shows how to use Firecase to create instant, reproducible development environments. Spin up a pre-configured workspace, share it with your team, and tear it down when you're done.

Why Firecase for Dev Environments?

Traditional dev environment approaches have trade-offs:

Approach	Startup	Isolation	Reproducibility	Cost
Local machine	Instant	None	Poor	Free
Docker	Seconds	Process-level	Good	Free
Cloud VMs (EC2, etc.)	Minutes	Full	Medium	$$$
Firecase	~150ms	Hardware-level	Excellent	Pay per use

Firecase gives you full VM isolation with container-like speed, plus persistent state and instant forking.

Basic Setup

1. Create an Instance from a Base Image

Start with a pre-built base image that has your language runtime and tools:

Bash

curl -X POST https://api.firecase.ai/instances \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "dev-workspace",
    "base_image_id": "550e8400-e29b-41d4-a716-446655440000",
    "vcpus": 4,
    "memory_mib": 4096
  }'

2. Start the VM

Bash

curl -X POST https://api.firecase.ai/instances/{id}/vm \
  -H "Authorization: Bearer $API_KEY"

3. Install Your Project Dependencies

Bash

# Clone your repo
curl -X POST https://api.firecase.ai/instances/{id}/exec \
  -H "Authorization: Bearer $API_KEY" \
  -d '{
    "command": ["git", "clone", "https://github.com/your-org/your-repo.git", "/home/user/project"],
    "timeout_ms": 60000
  }'

# Install dependencies
curl -X POST https://api.firecase.ai/instances/{id}/exec \
  -H "Authorization: Bearer $API_KEY" \
  -d '{
    "command": ["bash", "-c", "cd /home/user/project && npm install"],
    "timeout_ms": 120000
  }'

4. Checkpoint the Environment

Save this state so you can instantly restore or share it:

Bash

curl -X POST https://api.firecase.ai/instances/{id}/checkpoints \
  -H "Authorization: Bearer $API_KEY" \
  -d '{"label": "deps-installed"}'

5. Start Your Dev Server and Forward the Port

Bash

# Start the dev server in the background
curl -X POST https://api.firecase.ai/instances/{id}/exec \
  -H "Authorization: Bearer $API_KEY" \
  -d '{
    "command": ["bash", "-c", "cd /home/user/project && nohup npm run dev > /tmp/dev.log 2>&1 &"],
    "timeout_ms": 5000
  }'

# Forward port 3000
curl -X POST https://api.firecase.ai/instances/{id}/port-forwards \
  -H "Authorization: Bearer $API_KEY" \
  -d '{"guest_port": 3000, "label": "dev-server"}'

The response gives you a public URL to access your dev server.

SSH Access

For interactive development, configure SSH and use your local editor's remote extension (VS Code Remote SSH, JetBrains Gateway, etc.):

Bash

# Set your SSH key
curl -X PUT https://api.firecase.ai/instances/{id}/ssh-keys \
  -H "Authorization: Bearer $API_KEY" \
  -d '{"keys": ["ssh-ed25519 AAAAC3... user@laptop"]}'

# Get connection info
curl https://api.firecase.ai/instances/{id}/ssh \
  -H "Authorization: Bearer $API_KEY"

Then connect:

Bash

ssh root@compute-1.firecase.ai

Environment Templates

Create a reusable template by setting up an environment once, checkpointing it, and forking for each developer:

Python

# One-time setup: create the template
template = client.instances.create(name="template-nodejs-20")
client.vms.start(template.id)

# Install everything
client.exec(template.id, command=["bash", "-c", """
  curl -fsSL https://deb.nodesource.com/setup_20.x | bash -
  apt-get install -y nodejs
  npm install -g typescript tsx prettier eslint
"""])

# Checkpoint
checkpoint = client.checkpoints.create(template.id, label="nodejs-20-ready")
client.vms.stop(template.id)

# For each developer, fork from the template
def create_dev_env(developer_name: str):
    env = client.instances.fork(
        template.id,
        name=f"dev-{developer_name}",
        checkpoint_id=checkpoint.id
    )
    client.vms.start(env.id)
    return env

Each fork is instant and shares the base chunks. Developers only pay for storage of their changes.

Multi-Service Development

For microservice projects, spin up multiple sandboxes and connect them:

Python

# Backend API
api = client.instances.create(name="dev-api", vcpus=2, memory_mib=2048)
client.vms.start(api.id)
client.exec(api.id, command=["bash", "-c", "cd /app && npm start &"])
api_forward = client.port_forwards.create(api.id, guest_port=8080, label="api")

# Frontend
frontend = client.instances.create(name="dev-frontend", vcpus=2, memory_mib=2048)
client.vms.start(frontend.id)

# Point frontend at the API
client.exec(frontend.id, command=["bash", "-c",
    f"echo 'API_URL=http://{api_forward.hostname}:{api_forward.host_port}' > /app/.env"
])
client.exec(frontend.id, command=["bash", "-c", "cd /app && npm run dev &"])
frontend_forward = client.port_forwards.create(frontend.id, guest_port=3000, label="frontend")

print(f"Frontend: http://{frontend_forward.hostname}:{frontend_forward.host_port}")
print(f"API: http://{api_forward.hostname}:{api_forward.host_port}")

Cleanup

Stop and optionally delete environments when done:

Bash

# Stop (preserves state for later)
curl -X DELETE https://api.firecase.ai/instances/{id}/vm \
  -H "Authorization: Bearer $API_KEY"

# Delete (permanent)
curl -X DELETE https://api.firecase.ai/instances/{id} \
  -H "Authorization: Bearer $API_KEY"

Use auto_delete_on_stop: true in your profile to automatically clean up instances when the VM stops.

Next Steps

Learn about Custom Base Images to build your own templates
See Port Forwarding for exposing services
Explore SSH access for remote editor integration

AI Code Execution

Custom Base Images