How To Set Up An Ai Agent

Step 1 — Name the job before naming the tool

This guide is a decision tree. In ten minutes you’ll know which of four paths fits your problem, roughly what it costs, and what to read next. Written April 2026, against the current generation of tools (Claude Opus 4.7, GPT-5 class models, OpenAI AgentKit, Claude Agent SDK 0.1.48).

Step 2 — Pick the right lane

There are four real options in 2026. Pick the highest-level one that can do the job. Going lower buys flexibility but costs weeks of your time.

Lane A — Use a hosted agent product (1 hour)

Log every run: inputs, tool calls, outputs, token spend, user feedback. The best agents in production look boring — a narrow task, a tight tool set, 95%+ success, a human reviewing the 5%. The worst look impressive in demos and explode the first time they meet a real user.

Lane B — Use a no-code platform (1 day)

Lane C — Use an agent SDK (1 week)

Lane D — Use a framework (2+ weeks)

Step 3 — Decide where it runs

Step 4 — Wire it to the tools it needs

Step 5 — Start with read-only, add writes slowly

Step 6 — Budget for tokens and cap them

Step 7 — Measure and iterate

The honest shortlist