Running your LLM agent safely: Hands-on with Docker Sandboxes

Dan Ndombe
Staff Developer Success Advocate at Docker
Running agents safely with Docker Sandboxes
The moment your LLM stops just answering questions and starts acting — calling tools, writing files, hitting APIs, making decisions on its own — you need somewhere safe to let it do that. An agent with unrestricted access to your systems is an agent with an unrestricted attack surface, and that’s exactly the gap this workshop closes.
About this webinar
This hands-on session covers running LLM agents safely using Docker Sandboxes. Over the course of the workshop, Docker’s Dan Ndombe walks through sandboxing agents in practice — a code-along workshop rather than a slide deck. No prior Docker experience is required.
What you will learn
- What Docker Sandboxes are and why your LLM agent needs one
- How to spin up a sandboxed agent runtime using Docker Sandboxes (SBX)
- How to assign a real agent a task and watch it execute inside the sandbox
- How to tighten policies and permissions around agent behavior
- A reusable pattern for deploying this approach in your own AI and agent projects
Why agent sandboxing matters
Once you understand how large language models work, the next logical question is where you run them. When an agent can execute code, browse the web, or touch your file system, “where” is no longer a deployment detail — it’s a security decision. Docker Sandboxes give agents an isolated runtime with clearly defined boundaries: what they can access, what they can execute, and what they absolutely cannot touch.
Who should attend
Software engineers building or prototyping LLM-powered agents, DevOps and platform engineers responsible for how AI workloads get deployed, security-minded developers who want practical guardrails, and technical leads evaluating how to safely scale agentic AI initiatives. No prior Docker experience is required.
Dan Ndombe is a Staff Developer Success Advocate at Docker, helping developers build and ship software faster. He is a two-time founder and former engineer-turned-product manager with experience across companies including Netflix, Pinterest, and Calm.