From 5205abb55caa08e4ef8b4d69858740f1e4b46928 Mon Sep 17 00:00:00 2001
From: Jeremiah Lowin <153965+jlowin@users.noreply.github.com>
Date: Thu, 13 Jun 2024 23:18:03 -0400
Subject: [PATCH] minor updates

---
 docs/introduction.mdx | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/docs/introduction.mdx b/docs/introduction.mdx
index 18840001..3fe7bc27 100644
--- a/docs/introduction.mdx
+++ b/docs/introduction.mdx
@@ -85,17 +85,16 @@ This belief leads to three core design principles that underpin ControlFlow's ar
 
 ### 🛠️ Simple over complex
 
-ControlFlow believes in using **specialized, single-purpose LLMs** rather than monolithic models that try to do everything. Single-serving LLMs are more effective and efficient, ensuring that each task is handled by the right tool, leading to higher-quality results and easier management.
+ControlFlow believes in deploying specialized, single-purpose LLMs to a series of small problems, rather than using monolithic models that try to do everything. Single-serving LLMs are more effective and efficient, ensuring that each task is handled by the right tool, leading to higher-quality results and easier management.
 
 ### 🎯 Outcome over process
 
-ControlFlow advocates for a **declarative approach to defining AI workflows**. By focusing on desired outcomes instead of attempting to steer every action of the LLM, developers can create more predictable and controllable workflows, making it easier to achieve the desired results.
+ControlFlow takes a declarative approach to defining AI workflows. By focusing on outcomes instead of attempting to steer every action and decision of the LLM, developers can create more predictable and controllable workflows, making it easier to achieve the desired results.
 
 
 ### 🎛️ Explicit over implicit
 
-ControlFlow emphasizes the importance of treating LLMs as discrete, traditional software calls rather than relying on an end-to-end magic process. Explicitly defining tasks and checkpoints within a workflow ensures reliable results, allowing developers to maintain visibility and control while still leveraging the autonomy of LLMs.
-
+ControlFlow treats LLMs like discrete components of a software workflow rather than an end-to-end magic process. By explicitly defining tasks and checkpoints within workflows, developers can achieve reliable, predictable, and testable results. By focusing on code first and chat second, ControlFlow ensures that the artifacts of your workflows are structured data, not conversational text
 
 ## Why ControlFlow?