Uploaded January 2026 | Updated September 2026, 1 week ago
What happens when your AI coding agent gets a little too ambitious?
In this episode of AI Guide to the Galaxy, host Oleg Šelajev chats with Docker engineer Kevin Wittek about the rapidly growing need to sandbox AI agents. As AI tools become more autonomous, the risks of letting them run loose on your dev environment are very real (yes, even rm -rf ~ real).
Kevin shares how Docker’s approach (using microVMs and container-based sandboxes) helps developers run agents in a secure, isolated environment without losing productivity or control.
Topics include:
-- What “sandboxing” really means for AI agents
-- Why traditional OS-level protections (like Seatbelt) fall short
-- How Docker’s microVM-based sandbox lets agents run safely in “YOLO mode”
-- Secret injection, network firewalls, and filesystem isolation
-- The future of autonomous agents in software development
Whether you're building AI-powered developer tools or just curious how to keep your system safe from rogue AI behavior, this is a must-watch.
Subscribe for more conversations at the intersection of AI and developer tools
🔥 Want More Docker Content?
If you found this demo exciting, hit that like button and subscribe for more! We’ve got even more Docker demos coming your way in this ongoing series showcasing new tools, integrations, and powerful workflows to level up your projects. Stay tuned!
Where to find Docker:
Docker: docker.com
LinkedIn: linkedin.com/company/docker
Bluesky: https://bsky.app/profile/docker.com
X: @docker
Instagram: @dockerinc
#Docker #AI #Sandboxing #AgenticAI #LLM #DevTools #MicroVM #aiagents #aiguidetothegalaxy #aiguide
What happens when your AI coding agent gets a little too ambitious?
In this episode of AI Guide to the Galaxy, host Oleg Šelajev chats with Docker engineer Kevin Wittek about the rapidly growing need to sandbox AI agents. As AI tools become more autonomous, the risks of letting them run loose on your dev environment are very real (yes, even rm -rf ~ real).
Kevin shares how Docker’s approach (using microVMs and container-based sandboxes) helps developers run agents in a secure, isolated environment without losing productivity or control.
Topics include:
-- What “sandboxing” really means for AI agents
-- Why traditional OS-level protections (like Seatbelt) fall short
-- How Docker’s microVM-based sandbox lets agents run safely in “YOLO mode”
-- Secret injection, network firewalls, and filesystem isolation
-- The future of autonomous agents in software development
Whether you're building AI-powered developer tools or just curious how to keep your system safe from rogue AI behavior, this is a must-watch.
Subscribe for more conversations at the intersection of AI and developer tools
🔥 Want More Docker Content?
If you found this demo exciting, hit that like button and subscribe for more! We’ve got even more Docker demos coming your way in this ongoing series showcasing new tools, integrations, and powerful workflows to level up your projects. Stay tuned!
Where to find Docker:
Docker: docker.com
LinkedIn: linkedin.com/company/docker
Bluesky: https://bsky.app/profile/docker.com
X: @docker
Instagram: @dockerinc
#Docker #AI #Sandboxing #AgenticAI #LLM #DevTools #MicroVM #aiagents #aiguidetothegalaxy #aiguide
