Agent Substrate: Sandboxing is Not Enough
Running AI agents at scale isn’t just about getting them to work. You need isolation, fast startup, efficient resource utilization, and a way to run lots of agents concurrently without wasting cluster capacity.
Sandboxing is only the beginning.
Enter Agent Substrate, a new open source project designed to make running agentic workloads on Kubernetes more efficient and scalable while maintaining strong security and isolation.
In this hands-on workshop with Christian Posta and Lin Sun, you’ll go under the hood of Agent Substrate to understand its architecture, see how it efficiently packs more agents onto your infrastructure, and learn how it provides isolated sandboxes for running AI agents.
By the end of the workshop, you’ll know how to:
- Sandbox AI agents for stronger isolation and safer execution
- Pack more agents onto Kubernetes by using resources more efficiently
- Install and run Agent Substrate in a Kubernetes environment
- Understand the architecture, including WorkerPools, Actors, and Workers
- Control agent traffic with ingress/egress policies for sandboxed agents
Through the hands-on workshop, you’ll see how Agent Substrate brings security, isolation, and resource efficiency together to create a scalable substrate for running AI agents on Kubernetes.
Running AI agents at scale isn’t just about getting them to work. You need isolation, fast startup, efficient resource utilization, and a way to run lots of agents concurrently without wasting cluster capacity.
Sandboxing is only the beginning.
Enter Agent Substrate, a new open source project designed to make running agentic workloads on Kubernetes more efficient and scalable while maintaining strong security and isolation.
In this hands-on workshop with Christian Posta and Lin Sun, you’ll go under the hood of Agent Substrate to understand its architecture, see how it efficiently packs more agents onto your infrastructure, and learn how it provides isolated sandboxes for running AI agents.
By the end of the workshop, you’ll know how to:
- Sandbox AI agents for stronger isolation and safer execution
- Pack more agents onto Kubernetes by using resources more efficiently
- Install and run Agent Substrate in a Kubernetes environment
- Understand the architecture, including WorkerPools, Actors, and Workers
- Control agent traffic with ingress/egress policies for sandboxed agents
Through the hands-on workshop, you’ll see how Agent Substrate brings security, isolation, and resource efficiency together to create a scalable substrate for running AI agents on Kubernetes.
Speakers
Lin Sun
Lin is the Head of Open Source at Solo.io, contributing full-time to the open-source community. She serves on the CNCF Technical Oversight Committee (TOC), is a CNCF Ambassador, and is a maintainer for Istio, kgateway, and kagent. An international speaker at tech conferences, Lin frequently blogs about gateways, service meshes, cloud-native connectivity, AI agents, and MCP. She is the author of "Sidecar-less Istio Explained", is currently writing "AI Agents in Kubernetes", and holds more than 200 patents.
Christian Posta
Christian Posta (@christianposta) is Global Field CTO at Solo.io supporting customers and end users in their adoption of cloud-native technologies. He is an author for Manning and O’Reilly publications, open source contributor, blogger and sought after speaker on Envoy Proxy and Kubernetes technologies. Prior to Solo.io, Chrisitian was a Chief Architect at Red Hat, FuseSource and held engineering positions at organizations like Wells Fargo, Apollo Group, Intel.
