I am so glad to hear that. Then you’ll really like the next few pieces on engineering Agentic AI systems as well.
- Specifications and Controls, and how they enforce controlled outcomes
- Runtime Guards
- Autonomy policy - as fallback governance. This decides the autonomy ceiling on what the system can never do because agent tools can sometimes go rogue and have over privileged access.
This piece realy made me think. Excellent connection to planning.
I am so glad to hear that. Then you’ll really like the next few pieces on engineering Agentic AI systems as well.
- Specifications and Controls, and how they enforce controlled outcomes
- Runtime Guards
- Autonomy policy - as fallback governance. This decides the autonomy ceiling on what the system can never do because agent tools can sometimes go rogue and have over privileged access.