Deployment with SRE
SRE supports a flexible deployment lifecycle, allowing you to move from local development to managed cloud or private infrastructure environments without rewriting your agent code.
How SRE Starts Up
Each time SRE runs, it follows a consistent boot process. This process allows you to configure once and scale with confidence.
The architecture reference describes this process in detail.
import { SRE } from 'smyth-runtime';
const sre = SRE.init({
Storage: { Connector: 'S3', Settings: { bucket: 'my-bucket' } },
Cache: { Connector: 'Redis', Settings: { url: 'redis://prod-cluster' } },
Vault: { Connector: 'HashicorpVault', Settings: { url: 'https://vault.company.com' } },
LLM: { Connector: 'OpenAI', Settings: { apiKey: '...' } },
});
await sre.ready();
Deployment Environments
SRE is designed to be portable and flexible. You can deploy your agents in different environments, depending on your needs.
| Mode | Infrastructure | Common Use Case | How to Access |
|---|---|---|---|
| Agent Cloud | SmythOS-managed | Production, staging, MVPs | Via Studio |
| Local Runtime | Your own device | Debugging, local development | SDK or Docker |
| Self-Hosted | Your infrastructure | Enterprise, regulated workloads | Custom deployment |
Agent Cloud Deployment
Agent Cloud is the fully-managed hosting service provided by SmythOS. It is the fastest path to a production-ready deployment.
Benefits include:
- Auto-scaling, production-grade infrastructure with HTTPS
- No setup or maintenance required
- Centralized logs and traces available in Studio
- Continuous deployment workflows from Studio
Deployment steps:
- Open your agent in Studio.
- Click the Deploy button.
- Select the Agent Cloud target.
- Set your agent version and release notes.
- Click Deploy.
Local Development Setup
Running SRE on your own machine is ideal for development, testing workflows, or working offline. The Getting Started guide walks you through this process.
- Run code directly using the SDK
- Use RAM for cache and Local for storage connectors
- See immediate output in your local logs
Self-Hosted and Enterprise
For complete control, run SRE on your own servers or cloud infrastructure. Configure it with production connectors for storage, secrets, logging, and compute. This approach fits organizations with enterprise or regulatory requirements.