Our standard sandbox is hosted in our own secure cloud environment, which is not exclusively tied to any single provider like AWS.
The evaluation sandbox you access is a multi-tenant instance that we manage.
This approach lets your team test core functionality immediately, with zero setup required on your side.
For a dedicated proof-of-concept, we can deploy a single-tenant sandbox instance directly into your own AWS, Azure, or GCP environment. This flexibility is central to the design of OpenController by Lyzr, the control plane for enterprise AI agents.
A control plane is the neutral layer that governs everything beneath it. For AI agents, OpenController is that layer, built to be infrastructure-agnostic from the ground up.
Sandbox Options Reflect Production Reality
The default, multi-tenant sandbox is for rapid evaluation. It's the fastest way to see how OpenController discovers, observes, and governs agents.

When security and integration testing are the priority, a dedicated sandbox in your own VPC is the next step.
This isolates your evaluation completely.
It allows you to test agent integrations with your actual internal systems and data sources, all within your established network security perimeter. Your data never enters our environment.
The Path to Production is Your Own
OpenController is not a SaaS-only product. The deployment model for your evaluation mirrors your options for production.

You own the deployment strategy.
You can run OpenController in a dedicated VPC within your preferred cloud provider.
Or, for maximum control, you can deploy it entirely on-premise in your own data centers.
This ensures that all agent orchestration, metadata, and observability logs remain within your network boundary. It also means the control plane integrates cleanly with your existing CI/CD pipelines for deploying agents across the development, staging, and production environments you manage.
Conclusion: Deployment Follows Your Architecture
The key takeaway is this: OpenController adapts to your infrastructure, not the other way around. Whether in our cloud, your cloud, or your data center, the control plane is designed to fit your enterprise standards.
Ready to see how the control plane fits your architecture? Let's talk.
Related questions
Want to see how it fits your stack? Talk to us about OpenController by Lyzr.
Got a use case in mind?
8 weeks from use case to agents running in production.
Platform, people and FDEs, all in. Bring your environment. We'll co-build and stay until it's live.
Talk to UsFree resource
101 AI Agents Use Cases
100+ production-ready AI agent use cases across Marketing, Sales, HR, Banking, Insurance, Operations and a lot more
Download now