9 Building Blocks of Agentic AI

Pattern · Control Flow

Plan → act → observe → reflect — repeated until a goal is met. The base unit of agentic behaviour.

Plan

Act

Observe

Reflect

Protocol · Tool Integration

One interface, every tool. Standardised connectors instead of N×M custom integrations.

Architecture · Multi-Agent

Delegated reasoning, isolated context. An orchestrator splits work across specialised subagents and merges the result.

Infrastructure · Control Plane

One control plane, many models. Auth, rate-limits, logging, fallback — centralised in front of every provider.

Economics · Cost Model

Tokens are the unit of cost. Cache-hits land at ~10 % of the price; misses pay full freight.

Quality · Measurement

What you can’t measure, you can’t ship. Test cases drive model output through a judge — pass/fail aggregates into shipping signals.

Safety · Brand Protection

Input filter blocks PII and prompt injections; output filter blocks jailbreaks, leaks, and policy violations.

Operations · Visibility

You can’t fix what you can’t see. Traces, logs, and metrics from every agent step land on one dashboard.

Principle · Sutton, 2019

Handcrafted rules are brittle and lose long-term. Methods that scale with compute & data win — every time the cycle repeats.