Uploaded April 2026 | Updated September 2026, 2 weeks ago

---

This session with Michael Irwin, Principal Software Engineer at Docker, shows developers how to secure AI coding agents that bypass sandboxes, leak credentials, and delete filesystems. Using Docker Sandboxes and the MCP Toolkit, you’ll explore real attack scenarios and the guardrails Docker is building to give agents full power with safety.

In this session, you’ll learn to:
- Identify and block common agent vulnerabilities, including sandbox bypasses, API token leaks, and prompt injections.
- Use Docker Sandboxes and the MCP Toolkit to add guardrails and observability to agentic workflows.

Speaker:
Michael Irwin | Principal Engineer | Docker : linkedin.com/in/mikesir87
______
Learn data science, AI, and machine learning through our hands-on training programs: youtube.com/@Datasciencedojo/courses

Check our community webinars in this playlist: youtube.com/playlist?list=PL8eNk_zTBST-EBv2LDSW9Wx_V4Gy5OPFT

Check our latest Future of Data and AI Conference: youtube.com/playlist?list=PL8eNk_zTBST94qbEbiHeg3nSjncRDsrv8

Subscribe to our newsletter for data science content & infographics: datasciencedojo.com/newsletter

Love podcasts? Check out our Future of Data and AI Podcast with industry-expert guests: youtube.com/playlist?list=PL8eNk_zTBST_jMlmiokwBVfS_BqbAt0z2