About

This blog is by the team at Celesto AI. We write about deploying AI agents safely — sandboxing, security, observability, and the infrastructure decisions that come with putting agents into production.

Most of what we publish falls into a few categories:

• Agent security — how to isolate agents, handle untrusted input, and avoid common mistakes
• Infrastructure — VMs, microVMs, containers, and what actually works for agent workloads
• Tutorials — step-by-step guides for building and deploying agents
• Product updates — new features and changes to the Celesto platform

What is Celesto?

Celesto is a platform for deploying long-running AI agents and MCP servers. It handles the runtime infrastructure — sandboxing, tracing, and scaling — so you can focus on what your agent actually does.

If you want to learn more, check out the docs or the main site.

Community posts

We occasionally publish posts from external authors. If you have something interesting to share about AI agents, security, or infrastructure, reach out — we’d love to hear from you.

Get in touch

• Twitter/X: @celestoai
• Email: aniket@celesto.ai