Building effective agents

Analysis

Anthropic’s guide argues that effective LLM agents should start simple, use clear building blocks, and add autonomy only when simpler approaches fail. This matters because agents can improve performance on complex tasks but increase cost, latency, debugging difficulty, and the risk of compounding errors. The guide distinguishes predictable, code-defined workflows from agents that dynamically plan and use tools. It recommends first optimizing a single LLM call with retrieval and examples, then using workflows such as prompt chaining, routing, parallelization, orchestrator-workers, or evaluator-optimizer when their structure fits the task. Agents suit open-ended work with unpredictable steps, such as coding or computer use, especially when tools provide reliable feedback and humans can intervene. Developers should usually begin with direct LLM APIs rather than frameworks, because extra abstraction can hide prompts and responses. If they use frameworks, they should understand the underlying code. Tool design matters as much as prompt design: tools should have clear descriptions, natural formats, examples, safeguards, and thorough testing. Customer support and coding are highlighted as strong applications because both combine conversation with action and offer measurable outcomes. Coding agents can use automated tests for feedback, but human review remains necessary. The recommended approach is to measure results, preserve simplicity and transparency, document and test tools carefully, and increase system complexity only when it demonstrably improves outcomes.
  • The guide treats tool and interface design as a major reliability lever: Anthropic says it spent more effort optimizing tools than the overall prompt for its SWE-bench agent.
  • Its practical decision rule is to match architecture to task structure: fixed, measurable work favors workflows, while unpredictable work justifies agents with checkpoints, testing, and human review.

Build this article’s timeline

Extract dated events, key actors, and open questions using the same timeline analysis available in Signal.

Log in to build

Timeline generation is available to signed-in community members.

Five Ws 🤖 4 AI prefills · 1 still unanswered
Who
Anthropic and AWS
🤖 AI prefill
What
Anthropic’s guide argues that effective LLM agents should start simple, use clear building blocks, and add autonomy only when simpler approaches fail.
🤖 AI prefill
When
December 2024
🤖 AI prefill
Where Where did this occur?
Why
Centers on agentic systems, LLM agents, and workflows
🤖 AI prefill

Login to answer and earn XP for each gap you fill.

Agentic Systems Concepts agents, workflows, autonomy, LLMs, tool usage Agent Design Strategy simplicity, complexity, latency, cost, performance Augmented LLMs retrieval, tools, memory, search queries, Model Context Protocol Workflow Patterns prompt chaining, routing, parallelization, orchestrator-workers, evaluator-optimizer Production Agent Use reasoning, planning, error recovery, human interaction, complex inputs
agentic systems100% LLM agents98% workflows96% augmented LLM95% dynamic tool usage91% Model Context Protocol90% prompt chaining89% orchestrator-workers workflow88% evaluator-optimizer workflow87% parallelization workflow85%
Organizations: Anthropic2 AWS
5
Manipulation Score: Low How much this article uses emotional framing vs. straight reporting
Loaded Language 425%
Absolutist 315%
Threat/Panic 525%
Us vs Them 115%
Engagement Bait 220%
Manipulation score look right?

Factual Claims (5)

low technology
Anthropic recommends starting with simple, composable LLM patterns and adding complexity only when simpler approaches fail.
we recommend finding the simplest solution possible, and only increasing complexity when needed
medium technology
Agentic systems can improve task performance but trade off higher cost, latency, debugging difficulty, and potential compounding errors.
Agentic systems often trade latency and cost for better task performance
low technology
Workflows use predefined code paths, while agents dynamically direct their own processes and tool usage.
Workflows are systems where LLMs and tools are orchestrated through predefined code paths. Agents ... dynamically direct their own processes
low technology
Anthropic recommends optimizing a single LLM call with retrieval and examples before adopting multi-step workflows.
For many applications, however, optimizing single LLM calls with retrieval and in-context examples is usually enough.
medium technology
Agents are best suited to open-ended tasks such as coding or computer use when tools provide environmental feedback and humans can intervene.
Agents can be used for open-ended problems where it’s difficult or impossible to predict the required number of steps

Comments

Login to view and post comments.

🏠Portal 📰Links ❓Q&A 📅Events 💼Jobs