AI PM Ready

AI agents

From reactive chatbots to autonomous systems. What every AI PM must own: trust radius, tool governance, eval metrics, and when not to build an agent at all.

Agent vs Workflow vs Chatbot — When to Use What
Chatbot
Single-turn Q&A
Defined intents, finite answers
Real-time sub-second response
Low task complexity

Avoid when: When the task has more than 2 steps or requires external tool calls.

Workflow
Steps are known in advance
Deterministic and repeatable
Compliance or audit trail required
Lower cost per execution is priority

Avoid when: When the path through steps varies based on real-time conditions.

Agent
Steps unknown until runtime
Requires reasoning to plan a path
Multiple external tools needed
Must handle ambiguity and edge cases

Avoid when: When a deterministic workflow can do the job — agents cost 10x more and fail more often.

The Agentic Loop — A Real Trace
Step 01 — Task Received
User: "Book me on the cheapest NYC→London flight tomorrow."
What happens

The orchestrator receives the goal and must plan: what tools exist, what order to call them, what confirmation is needed before any action executes.

PM owns this

You define the goal format and ambiguity handling. What happens if "tomorrow" is ambiguous across timezones? That is a PM spec decision, not an engineering edge case.

Levels of Autonomy — Select a Level
L2 — Human Approves
AI executes after explicit human confirmation.
Real example

Agent drafts an email and surfaces it for review. You click Send. The agent cannot send without your approval.

Use when

Financial actions, external communications, and any operation that cannot be undone.

The Trust Radius — the most important mental model for agent PMs: Every agent has a radius of actions it can take autonomously. PMs define that radius — not engineers. Too small and the agent delivers no value. Too large and you get catastrophic failures. Every agent feature ships with a trust radius doc: what it can do freely, what requires confirmation, what is permanently blocked.

Try It Yourself
Loop trace

Act as a travel booking agent. You have access to: search_flights, book_flight, cancel_booking, send_email. I want to change my NYC-London flight to tomorrow. Walk me through every decision you make and every tool call, including what could go wrong at each step and what confirmation you'd ask for.

Trust radius

I'm a PM at a fintech company. My agent can read transactions and send payment reminders. Write the trust radius document for this agent: what it can do autonomously, what needs human approval, and what is permanently prohibited. Be specific about each action type.

Agent vs workflow

I want to automate customer onboarding. Steps: verify ID, check credit score, assign account manager, send welcome email. Should I build an agent or a deterministic workflow? Walk me through the exact decision criteria and give me your recommendation.

Key Takeaways
Agent ≠ chatbot ≠ workflow

Agents are for unknown step paths and multi-tool reasoning. Default to the simpler option — a deterministic workflow is cheaper, more reliable, and easier to debug.

Loops are probabilistic, not deterministic

You cannot write acceptance criteria like a traditional feature. Define failure modes and stopping conditions, not just success paths.

Trust Radius is your most important PM artifact

Classify every action: runs freely, requires confirmation, or permanently blocked. This document belongs in the PRD, not the engineering design doc.

Partial completion is the worst failure state

An agent that completes 2 of 4 steps and fails is worse than one that never started. Write rollback specs for every write and irreversible action before engineering begins.

Was this page useful?