Hello everyone,
The new Code Execution feature on Vertex AI Agent Engine is NOW IN PREVIEW!
To help you build LLM apps like agents that run codes, Vertex AI Agent Engine is introducing Code Execution. This new managed service addresses the significant security risks and operational overhead of running LLM-generated code by providing a secure, managed runtime environment.
With this new feature, you’ll learn how to:
- Simplify your workflow: Eliminate the engineering challenge of building your own execution infrastructure by using a simple, fully managed API to create, execute, and manage runtimes.
- Integrated: Use Code Execution as a fundamental building block. You can call it directly via the SDK or integrate it as a tool with your LLMs and agent frameworks like the Agent Development Kit.
- Run Code safely: Run generated code in a hardened, isolated sandbox.
Check out the following resources to get started:
As always, I’d love to hear your feedback. And if you have questions, you can also connect with me here, on LinkedIn or X/Twitter.