Blog
Tutorials, news, and release notes.
Showing 1–12 of 16 articles
Enterprise AI7 min read
The Agent Harness Can Restart. Your Business Transaction Still Needs a Recovery Design.
Microsoft’s Foundry harness and long-running recovery change the agent runtime, but banks still need durable task state, scoped tool authority and idempotent external effects.
Enterprise AI4 min read
Enterprise AI Agent Control Plane: Architecture and Controls
OpenAI's dots make persistent agents tangible. Enterprise differentiation moves to identity, authority, observability and control across systems.
Enterprise AI4 min read
LLM Model Routing: Cost per Successful Workflow
New model pricing makes routing attractive, but the useful metric is cost per successful workflow rather than tokens alone.
Architecture7 min read
AI Software Delivery Governance: Release Evidence and Controls
Governing AI-generated software requires verified build inputs, scoped credentials, and release evidence tied to immutable artifacts.
Enterprise AI4 min read
Enterprise RAG Security: Permissions and Tenant Isolation
Why enterprise knowledge systems need identity-aware retrieval, document lineage, policy freshness and explicit handling of unknowns.
Enterprise AI3 min read
AI Workflow Orchestration: Beyond Agent Tool Calls
A customer-onboarding exception exposes the missing architecture between tool calls and accountable enterprise work.
Enterprise AI4 min read
AI Agent Governance: Identity, Authority and Accountability
An operational control model for enterprise agents: identity, authority, evidence, human decision gates and recovery.
Enterprise AI4 min read
Enterprise AI Workflows: State, Approvals and Recovery
Why tool-connected agents are not enough: durable state, orchestration, exception handling and accountable business outcomes.
Enterprise AI4 min read
Enterprise AI Total Cost: Cost per Business Outcome
A practical cost model for enterprise agents, including inference, orchestration, retrieval, operations and customer-hosted trade-offs.
Enterprise AI4 min read
AI Contract Review: Evidence and Human Approval
From clause extraction to policy evidence and an accountable decision: designing contract review as a controlled enterprise workflow.
Architecture7 min read
AI Agent Sandboxing: Enforcing Authority Boundaries
Independent runtime boundaries for enterprise agents: what Microsoft’s MXC release changes and where service-side controls remain essential.
Architecture6 min read
AI Agent Instruction Governance: Scope, Versions and Audit
Claude's new compliance visibility into saved instructions highlights the need to separate personalization, controlled configuration, and action authorization.