diff --git a/README.md b/README.md
index 3100a3ba..704b2650 100644
--- a/README.md
+++ b/README.md
@@ -28,7 +28,7 @@ _Arch is a smart proxy server designed as a modular edge and AI gateway for agen
 
 AI demos are easy to hack. But once you move past a prototype, you’re stuck building and maintaining low-level plumbing code that slows down real innovation. For example:
 
-- **Routing & orchestration.** When routing logic lives in code, you’re stuck in one of two modes: either you write and maintain it yourself, or you inherit whatever a framework bakes in. In both cases, you must keep routing consistent across agents which means pushing code changes everywhere. Slowing down iteration, risks drift between deployments, and makes experimenting with new policies a refactor instead of a config flip.
+- **Routing & orchestration.** Put routing in code and you’ve got two choices: maintain it yourself or live with a framework’s baked-in logic. Either way, keeping routing consistent means pushing code changes across all your agents and services, slowing iteration and turning every policy tweak into a refactor instead of a config flip.
 - **Model integration churn.** Frameworks wire LLM integrations directly into code abstractions, making it hard to add or swap models without touching application code — meaning you’ll have to do codewide search/replace every time you want to experiment with a new model or version.
 - **Observability & governance.** Logging, tracing, and guardrails are baked in as tightly coupled features, so bringing in best-of-breed solutions is painful and often requires digging through the guts of a framework.
 - **Prompt engineering overhead**. Input validation, clarifying vague user input, and coercing outputs into the right schema all pile up, turning what should be design work into low-level plumbing work.