Articles
How agents ship
26 guides on the worker, the eval, and the runtime.
21
How to Auto-Run Ops Runbooks With an AI Agent
Let an agent run incident and ops runbooks by making done machine-checkable, starting read-only, putting writes behind approval, and paging a human on miss
16 min · Aug 25, 2026
Aug 25, 2026
16 min read
22
What Is an AI Agent (Not a Chatbot)
An AI agent is software given a job, a short list of tools, and a finish line
16 min · Mar 12, 2026
Mar 12, 2026
16 min read
23
AI Agent vs Chatbot: How to Tell Which You Need
A chatbot replies. An AI agent writes to a system of record
16 min · Mar 28, 2026
Mar 28, 2026
16 min read
24
How to Put an AI Agent in Production Without Shipping a Demo
Five steps to put an AI agent in production: a finish line you can test, a short tool list, a runtime that can stop it, an eval on frozen jobs, then one real desk
17 min · Apr 2, 2026
Apr 2, 2026
17 min read
25
How to Evaluate AI Agents on the Artifact, Not the Chat
Score the artifact, not the chat. Freeze jobs you already know, run them more than once, log every tool call, and keep a baseline before anyone depends on the agent
16 min · Apr 18, 2026
Apr 18, 2026
16 min read
26
How to Run AI Agents Without a Human in Every Step
Unattended agents are a budget of tools, retries, and stop rules. People stay at the edge: goals, grants, and review
17 min · Jun 11, 2026
Jun 11, 2026
17 min read