Axonius uses Amazon Bedrock AgentCore to run AI agents across hundreds of isolated SaaS customer environments.

Axonius, an asset intelligence platform, has deployed its first AI agent on Amazon Bedrock AgentCore. The agent interprets the state of large enterprise environments, identifies gaps and risks, and analyzes millions of data points from dozens of concurrent integration sources.

Axonius runs its software as a service on AWS and manages hundreds of isolated customer environments. The company adopted Amazon Bedrock AgentCore to add agentic workloads while keeping its existing tenant management methodology.

For builders, the key choice is the multi-tenancy pattern: silo, pool, or bridge. The silo model gives each tenant a dedicated agent, while the pool model shares one agent and isolates user sessions with unique session IDs. The bridge model mixes dedicated agents with shared resources such as Amazon Bedrock Knowledge Bases.

Axonius’s first agent lets junior analysts run complex analyses without tying up senior analysts in hours of manual work. The company now needs to manage cost tracking and security at the tenant level as it expands agent workloads to more customers.

What matters

• Axonius built its first AI agent for asset intelligence using Amazon Bedrock AgentCore.
• Builders can isolate each user session with a unique session ID when using the agent pool model.
• Axonius needs to track agent costs per customer as it scales multi-tenant workloads.

Why it matters

Axonius needs to track agent costs per customer as it scales multi-tenant workloads.

This GenAI News article was prepared in original wording using reporting and materials published by AWS Machine Learning Blog. Source reference: https://aws.amazon.com/blogs/machine-learning/how-axonius-built-secure-multi-tenant-ai-agents-on-bedrock-agentcore/.

Drafted by the GenAI News review pipeline.