diff --git a/site/intro.md b/site/intro.md
index b66b3f6c..de61cfd3 100644
--- a/site/intro.md
+++ b/site/intro.md
@@ -1,15 +1,15 @@
-# TinyTorch: Build ML Systems from Scratch
+# TinyTorch: Tensors to Systems
Don't just import it. Build it.
-Build a complete neural network framework from scratch—tensors to transformers—and understand how ML systems really work.
+Build a complete ML framework from tensors to systems—understand how PyTorch, TensorFlow, and JAX really work.
@@ -42,35 +42,13 @@ Build ML systems understanding through progressive tiers—from mathematical fou
-Each tier unlocks **historical milestones** where you validate implementations by recreating breakthrough moments in ML history. **[View milestone demos](chapters/milestones.html)** to see what you'll build.
+**[Complete course structure](chapters/00-introduction)** • **[Daily workflow guide](student-workflow)** • **[Join the community](community)**
-**[Complete course structure](chapters/00-introduction.html)** • **[Daily workflow guide](student-workflow.html)** • **[Join the community](community.html)**
+## Validation Through Milestones
-## Recreate 70 Years of ML History
+Validate your implementations with concrete benchmarks—MNIST accuracy, CIFAR-10 performance, transformer text generation. Each milestone proves your code works.
-As you progress through tiers, you'll validate your work by recreating historically significant breakthroughs:
-
-
-
-
-
-**1957 — Perceptron** • First trainable neural network
-
-**1969 — XOR Crisis** • Solved with hidden layers
-
-**1986 — Backpropagation** • Multi-layer learning
-
-**1998 — CNNs** • Spatial intelligence
-
-**2017 — Transformers** • Attention mechanisms
-
-**2018 — Systems** • Performance optimization
-
-
-
-