These labs are about building Agents and running them on Kubernetes: Models, MCP tools, an AI Gateway in front of the traffic, evals, and how much autonomy you hand over.
The Platform Engineering assistant has a goal of helping DevOps, Platform Engineers, and SREs find a solid balance between helping with implementation from an autonomy perspective. The important thing to keep in mind is there are different levels on autonomy ranging from just about zero control of what the Agent can access and the actions it can perform to an Agent having full control of an environment.
Hardware decisions for running Models locally (coming soon)
There's nothing like waking up at 2:00 AM to an alert and having to stay awake for 1.5 hours looking through logs with no end in sight to actually figuring out the issue. On the opposite, yet same spectrum, what if you're deploying a new application stack or piece of infrastructure and something goes wrong? Your first thought is to go look at logs, traces, metrics, only to be bogged down by it.
The autonomous k8s engineer takes autonomy to a whole other level. It can figure out what is going on, fix the issue, deploy what is needed, and ensure the environment is running smooth all without you having to even know about it.