Skip to content
back to the archive page
#PlatformEngineering

3 AImigo S1E13: Infrastructure for Agentic Development

You leave a few tasks for agents in the evening. In the morning you have changes, test results, and new questions. Who will check the result, approve the merge, and stop a dangerous action? What if there are thousands of agents?

On Friday at noon Moscow time, in episode 13, we'll discuss how development infrastructure changes under the load agents create. We'll follow a change from task definition through release:

– How do you control the flow of tasks and preserve history when an agent fails?
– What happens to CI, version control, and merge queues?
– How do you verify the result and the verifier?
– How do you grant and revoke agent permissions?
– How do you calculate the cost of an accepted change once failed attempts, checks, and rollbacks are included?

We'll compare approaches from Anthropic, OpenAI, and other companies and discuss where your team can start.

Bring your own examples. Where is agentic development already hitting infrastructure limits for you?

#AI4SDLC #DevOps #DevEx #Infra #Software #Engineering #AI

Open video on YouTube