Logo
Search
Sign Up
Logo
Oliver Buchannon
Mariana Lungu
The Boundary Around the Action

Jul 22, 2026

•

2 min read

The Boundary Around the Action

Why AI systems need a defined blast radius before they act.

Mariana Lungu
Mariana Lungu
The System Needs a Way Back

Jul 15, 2026

•

3 min read

The System Needs a Way Back

Why AI actions need recovery paths, not just approval.

Mariana Lungu
Mariana Lungu
The Check Before the Action

Jul 8, 2026

•

2 min read

The Check Before the Action

Why AI systems need approval points before they change real systems.

Mariana Lungu
Mariana Lungu
The Record Behind the Answer

Jul 1, 2026

•

3 min read

The Record Behind the Answer

Why provenance becomes critical when AI starts taking action.

Mariana Lungu
Mariana Lungu
The Permission That Stayed

Jun 24, 2026

•

2 min read

The Permission That Stayed

Access that stays after the reason is gone.

Mariana Lungu
Mariana Lungu
The Work AI Still Cannot Do

Jun 17, 2026

•

2 min read

The Work AI Still Cannot Do

AI can move fast. Trust still has to be engineered.

Mariana Lungu
Mariana Lungu
Before You Trust the Output

Jun 11, 2026

•

2 min read

Before You Trust the Output

Data, access, and failure are where trust is either built or lost.

Mariana Lungu
Mariana Lungu
Why I’m Writing Beyond the Prompt

Jun 9, 2026

•

2 min read

Why I’m Writing Beyond the Prompt

AI changed the interface. It did not remove the need for engineering judgment.

Mariana Lungu
Mariana Lungu

Beyond the Prompt

Practical notes on secure data systems, AI-aware engineering, and the decisions behind reliable technical work.

© 2026 Archetype Core. Beyond the Prompt..
beehiivPowered by beehiiv