Blog
Notes from our engineering work: custom software, AI and automation, systems integration, and what it takes to keep software healthy in production.
Blog posts
by Scott Miller
AI Agents Need Execution Governance, Not Just Access Control
Enterprise security has spent decades answering an essential question: Who is allowed to access this system? Identity and access management systems answer that question by authenticating users, assigning roles, enforcing permissions, and recording access. Those controls remain necessary as organizations introduce AI agents into operational environments.
by Scott Miller
Should an AI agent inherit a human's shell access?
Organizations are beginning to connect AI agents to the same administrative tools their engineers use.
- AIGovernance
- Cybersecurity
- DevOps
- IdentitySecurity
- SRE
by Scott Miller
Why audit logs do not prove operational outcomes
An audit log can tell us that an action was requested, authorized, and submitted. It cannot always tell us whether the intended operational result actually occurred.
- AIGovernance
- AIInfrastructure
- Cybersecurity
- DevOps
- SRE
by Scott Miller
The AI agent said it succeeded, But did it!
Imagine an AI agent is asked to restart a production service. The agent calls the appropriate tool. The command executes....
- AgenticAI
- AI
- AIGovernance
- Cybersecurity
- DevOps
- EnterpriseArchitecture
- PlatformEngineering
- SRE
by Scott Miller
Authorization isn't Execution
As enterprises give AI agents more authority to perform real operational work, I think we need to distinguish between two very different questions...
- AgenticAI
- AI
- AIGovernance
- Cybersecurity
- DevOps
- EnterpriseArchitecture
- PlatformEngineering
by Scott Miller
Rethinking Infrastructure for AI
We're spending enormous effort making AI models smarter. But are we spending enough time thinking about the infrastructure we're asking them to operate?
- AIAgents
- AIGovernance
- DevOps
- InfrastructureAutomation
- PlatformEngineering
- ResourceShell
- SRE
by Scott Miller
Why AI Agents Shouldn't Execute Shell Commands
As AI becomes more involved in operating infrastructure, there's a question I believe deserves more attention:
- AIAgents
- AIGovernance
- DevOps
- InfrastructureAutomation
- OpenSource
- PlatformEngineering
- ResourceShell
Have a project that needs this kind of thinking?
Tell us what you're building or what needs fixing, and we'll help you work out the smallest useful first step.