Workflows are not agents: the distinction that prevents most failed agent projects
- What it is
- The post by Erik Schluntz and Barry Zhang at Anthropic separates workflows (predefined code paths that call models) from agents (the model decides its own path), catalogs five workflow patterns — chaining, routing, parallelization, orchestrator-workers, evaluator-optimizer — and recommends starting with the simplest that solves the problem.
- What was demonstrated
- There is no benchmark; the result is a taxonomy. The central recommendation: use agents only when the problem needs flexibility and decisions that depend on what the model finds, and accept the cost and latency that come with it.
- What was not
- It is a vendor text, written for Anthropic's own customers. The examples are generic on purpose.
- Why it matters
- It is the vocabulary product and engineering teams need to share before discussing “putting an agent” on anything.
Appears in trails
Same theme
Agents→Tencent open-sourced a 770B MoE under Apache 2.0 — and it edges Kimi K3 on code with a quarter of the parameters
On real scientific software, the best coding agent solves fewer than half the bugs
ReAct: interleaving reasoning and action is what made the first language agents work
MCP became the tool standard for agents — and since December 2025 it no longer belongs to Anthropic
Get the next Cinco
Tuesday, 7am BRT, in your inbox. Five items, with sources. No daily newsletter, no promotions, and unsubscribing is one click.