Learn how Salesforce defined clear protection boundaries as Agentforce expanded, designed a layered access-control plane, and much more.

The Salesforce Engineering Blog offers a deep dive into Salesforce technologies, providing technical insights, best practices, and real-world examples. Developers can explore topics such as Salesforce development, architecture, and integrations, gaining  knowledge to build innovative solutions on the Salesforce platform. With contributions from Salesforce engineers and industry experts, the blog serves as a resource for the Salesforce community.

Salesforce Engineering

Salesforce's Developer Access team manages the security layer for Agentforce, handling over 11 million daily AI agent calls. The architecture implements a layered access-control plane with three core invariants: authentication validation at the platform edge, authorization enforced by the core platform, and end-to-end request integrity through Salesforce-minted tokens. Multi-tenant reliability is achieved through per-organization rate limits, Kubernetes horizontal autoscaling, and session-based tracking using in-memory distributed data stores. The team acts as a middle-layer API service between clients and downstream systems, ensuring consistent security enforcement while supporting rapid platform evolution.

How Agentforce Runs Secure AI Agents at 11 Million Calls Per Day