Category archive
An agent is a control loop with a durable state problem.
An agent demo needs one good trajectory. An agent in production needs a loop that can be resumed, a tool surface that can be authorised, and a run whose state survives the process that started it. These articles are about the second thing.
Articles in Agents
-
Engineering the Agent Tool Loop
A concrete architecture for tool selection, state transitions, permissions, budgets, and recovery in production agents.
-
Durable Execution for Agents That Outlive a Web Request
Persisting checkpoints, replaying deterministically, handling human approval, and surviving crashes without duplicating side effects.
-
Model Context Protocol in Production
Designing MCP servers, capability discovery, authorization, transport, schemas, and observability for real agent ecosystems.
Apply the category
Put this against a real system.
If a decision in Agents is in front of you right now, the fastest version of this is a call: bring the architecture, the failure you are seeing, and the constraint you cannot move.
Discuss the system
