# At A Glance (/docs/runtime/runtime-at-a-glance)



Runtime is the foundational layer of BullSequana AI. It provides the core services that make the rest of the platform operational: secure access, traffic exposure, inference execution, workflow engines, storage, observability, and delivery automation.

What Runtime Includes [#what-runtime-includes]

Runtime is organized around a small number of platform domains:

* **Network** for exposing and routing services
* **Access and security** for identity, authorization, secrets, and certificates
* **Inference** for running model-serving workloads
* **Workflows and automation** for event-driven and durable process execution
* **Data and storage** for operational data and object storage
* **Observability** for metrics, logs, traces, and dashboards
* **Delivery** for image publishing and GitOps deployment

Who Runtime Is For [#who-runtime-is-for]

Runtime matters to several audiences:

* **Customers and buyers**, who want a secure and production-ready base for AI and data services
* **Platform teams**, who operate the shared infrastructure layer
* **Solution architects**, who need to understand how higher layers depend on the platform
* **Developers**, who build on top of the services Runtime exposes

What Runtime Delivers [#what-runtime-delivers]

Runtime gives the platform a consistent way to:

* expose internal and external services
* authenticate users and services
* enforce authorization and secrets handling
* run inference engines and workflow systems
* persist application state and artifacts
* observe health, reliability, and performance
* deploy and update components through GitOps workflows

What Runtime Enables [#what-runtime-enables]

CoreAI and ProAI do not replace Runtime. They build on it.

Runtime is therefore the layer that turns infrastructure into an operational AI platform rather than a collection of disconnected services.