Anthropic
guidelineactive

Building Effective AI Agents

Anthropic

View original resource

Anthropic engineering write-up distinguishing deterministic workflows from agents and documenting composable patterns like prompt chaining, routing, parallelisation, orchestrator-workers, and evaluator-optimiser. Recommends starting with the simplest pattern and adding autonomy only where it pays off.

Tags

agentic AIfoundations

At a glance

Published

2024

Jurisdiction

Global

Category

Foundations

Access

Public access

Build your AI governance program

VerifyWise helps you implement AI governance frameworks, track compliance, and manage risk across your AI systems.

Building Effective AI Agents | VerifyWise AI Governance Library