Cloud Native Deep Dive

Many of us are "hands-on-keyboard" interacting with Agents in our terminal or via a desktop app.

A typical gateway comes with a lot of policy enforcement, authentication/authorization implementations, and filtering for both agentic and API-centric traffic.

As of 09/02/2026, the ActorTemplate object no longer exists.

The increase in cost of GPUs and AI is hitting an all-time high, so much so that it's hard to even get GPU usage approved within a public cloud environment.

The MCP protocol can now be stateful or stateless, ingest only one tool to the Agent, call API endpoints, and use functions/methods in a codebase as tools.

Without proper controls for how many times an Agent/LLM can hit an MCP Server, you open yourself up to potential DOS attacks, memory hogging, insane API bills, and server/system overload.

Three times over the past month, the industry has seen Agents break out of sandboxes.

In the majority of conversations around ML workflows, the same four problems come up: Long-running jobs that die halfway through and need to be restarted from scratch DAGs that can’t dynamically branch, loop, or call tools at runtime GPUs sitting over-pro…

With the stateless MCP spec now officially out as of July 28th, 2026, there are now two methods of connecting to and configuring an MCP Server.

Think about this scenario: you're working on a complex application refactor, and after the initial stages of performing said refactor, you decide to run tests and see if the updates worked as expected or if there are regressions.

Open Weight Models (Qwen, DeepSeek, Kimi) need a place to run that gives the ability to have shareable GPUs, a first-class orchestration/scheduler, and traffic routing capabilities that engineers are comfortable with.

Sandboxed Agents means we're going a level deeper in terms of where AI runs.

High Availability (HA) is the cornerstone of an efficient, scalable, performant system.

Cost in AI will vary per user and department.

Tokens turn into dollars, and the tokens are expanded upon in many different ways.

Cost is everything.

Putting kagent and Agent Substrate together, you have a fully isolated Agentic platform with a declarative runtime to interact with Agents running in a sandboxed environment, and the best part is you can interact with them like any other Agent.