diff --git a/modules/source/05_autograd/autograd_dev.ipynb b/modules/source/05_autograd/autograd_dev.ipynb
index 256ef10e..ca9e17fc 100644
--- a/modules/source/05_autograd/autograd_dev.ipynb
+++ b/modules/source/05_autograd/autograd_dev.ipynb
@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "markdown",
-   "id": "e6565b3a",
+   "id": "11a866a5",
    "metadata": {
     "cell_marker": "\"\"\""
    },
@@ -54,7 +54,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "49c7d7b9",
+   "id": "6f716656",
    "metadata": {
     "nbgrader": {
      "grade": false,
@@ -77,7 +77,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "f4f60314",
+   "id": "1c5fcfe6",
    "metadata": {
     "cell_marker": "\"\"\""
    },
@@ -131,7 +131,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "ed50cafc",
+   "id": "82cafe21",
    "metadata": {
     "cell_marker": "\"\"\""
    },
@@ -190,7 +190,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "6ba5f78e",
+   "id": "47bd67c9",
    "metadata": {
     "cell_marker": "\"\"\""
    },
@@ -227,7 +227,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "59b74ab0",
+   "id": "cce8538a",
    "metadata": {
     "cell_marker": "\"\"\"",
     "lines_to_next_cell": 1
@@ -255,7 +255,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "7cca5f0b",
+   "id": "7c604fa6",
    "metadata": {
     "lines_to_next_cell": 1,
     "nbgrader": {
@@ -321,7 +321,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "26284ed3",
+   "id": "f721b07e",
    "metadata": {
     "cell_marker": "\"\"\""
    },
@@ -360,7 +360,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "3e6303f0",
+   "id": "b783a909",
    "metadata": {
     "cell_marker": "\"\"\"",
     "lines_to_next_cell": 1
@@ -389,7 +389,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "f18f8fb1",
+   "id": "b8c92aa2",
    "metadata": {
     "lines_to_next_cell": 1,
     "nbgrader": {
@@ -444,7 +444,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "bb6389e0",
+   "id": "31a8a1ab",
    "metadata": {
     "cell_marker": "\"\"\"",
     "lines_to_next_cell": 1
@@ -477,7 +477,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "2b0a5371",
+   "id": "1a6762d0",
    "metadata": {
     "lines_to_next_cell": 1,
     "nbgrader": {
@@ -535,7 +535,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "4bcfb89b",
+   "id": "11567a68",
    "metadata": {
     "cell_marker": "\"\"\"",
     "lines_to_next_cell": 1
@@ -570,7 +570,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "74ead8e1",
+   "id": "102ba9f6",
    "metadata": {
     "lines_to_next_cell": 1,
     "nbgrader": {
@@ -627,7 +627,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "2f21a9f9",
+   "id": "d9496bda",
    "metadata": {
     "cell_marker": "\"\"\"",
     "lines_to_next_cell": 1
@@ -658,7 +658,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "a79f2e00",
+   "id": "37f9b250",
    "metadata": {
     "lines_to_next_cell": 1,
     "nbgrader": {
@@ -706,7 +706,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "c9759fa8",
+   "id": "116f71ea",
    "metadata": {
     "cell_marker": "\"\"\"",
     "lines_to_next_cell": 1
@@ -722,7 +722,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "17917493",
+   "id": "b2120ecf",
    "metadata": {
     "nbgrader": {
      "grade": true,
@@ -769,7 +769,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "ae3458ec",
+   "id": "9685115d",
    "metadata": {
     "cell_marker": "\"\"\""
    },
@@ -804,7 +804,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "4a2b4d65",
+   "id": "5612e207",
    "metadata": {
     "cell_marker": "\"\"\"",
     "lines_to_next_cell": 1
@@ -830,7 +830,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "9070836c",
+   "id": "b49922a3",
    "metadata": {
     "nbgrader": {
      "grade": false,
@@ -874,7 +874,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "1f26d8ca",
+   "id": "e79f5497",
    "metadata": {
     "nbgrader": {
      "grade": false,
@@ -918,7 +918,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "264226b7",
+   "id": "58b86487",
    "metadata": {
     "nbgrader": {
      "grade": false,
@@ -928,6 +928,7 @@
    },
    "outputs": [],
    "source": [
+    "#| export\n",
     "def enable_autograd():\n",
     "    \"\"\"\n",
     "    Enable gradient tracking for all Tensor operations.\n",
@@ -1125,7 +1126,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "a7fdf318",
+   "id": "d03e54f6",
    "metadata": {
     "cell_marker": "\"\"\"",
     "lines_to_next_cell": 1
@@ -1141,7 +1142,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "51b431ef",
+   "id": "1bae0903",
    "metadata": {
     "nbgrader": {
      "grade": true,
@@ -1189,7 +1190,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "5253c5dc",
+   "id": "fc159b24",
    "metadata": {
     "cell_marker": "\"\"\"",
     "lines_to_next_cell": 1
@@ -1203,7 +1204,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "dea69c7d",
+   "id": "92f51d47",
    "metadata": {
     "lines_to_next_cell": 1,
     "nbgrader": {
@@ -1316,7 +1317,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "9725e9c2",
+   "id": "ef3b1668",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1327,7 +1328,7 @@
   },
   {
    "cell_type": "markdown",
-   "id": "bd5d740f",
+   "id": "7728d17d",
    "metadata": {
     "cell_marker": "\"\"\""
    },
diff --git a/modules/source/05_autograd/autograd_dev.py b/modules/source/05_autograd/autograd_dev.py
index e50907be..d2d27cc1 100644
--- a/modules/source/05_autograd/autograd_dev.py
+++ b/modules/source/05_autograd/autograd_dev.py
@@ -735,6 +735,7 @@ class BCEBackward(Function):
 
 
 # %% nbgrader={"grade": false, "grade_id": "enable-autograd", "solution": true}
+#| export
 def enable_autograd():
     """
     Enable gradient tracking for all Tensor operations.
diff --git a/tinytorch/core/autograd.py b/tinytorch/core/autograd.py
index 7b2efb5a..14e18f35 100644
--- a/tinytorch/core/autograd.py
+++ b/tinytorch/core/autograd.py
@@ -1,7 +1,22 @@
-# AUTOGENERATED! DO NOT EDIT! File to edit: ../../modules/source/05_autograd/autograd_dev.ipynb.
-
+# ╔═══════════════════════════════════════════════════════════════════════════════╗
+# ║                        🚨 CRITICAL WARNING 🚨                                ║
+# ║                     AUTOGENERATED! DO NOT EDIT!                              ║
+# ║                                                                               ║
+# ║  This file is AUTOMATICALLY GENERATED from source modules.                   ║
+# ║  ANY CHANGES MADE HERE WILL BE LOST when modules are re-exported!            ║
+# ║                                                                               ║
+# ║  ✅ TO EDIT: modules/source/09_autograd/autograd_dev.py             ║
+# ║  ✅ TO EXPORT: Run 'tito module complete <module_name>'                      ║
+# ║                                                                               ║
+# ║  🛡️ STUDENT PROTECTION: This file contains optimized implementations.        ║
+# ║     Editing it directly may break module functionality and training.         ║
+# ║                                                                               ║
+# ║  🎓 LEARNING TIP: Work in modules/source/ - that's where real development    ║
+# ║     happens! The tinytorch/ directory is just the compiled output.           ║
+# ╚═══════════════════════════════════════════════════════════════════════════════╝
 # %% auto 0
-__all__ = ['Function', 'AddBackward', 'MulBackward', 'MatmulBackward', 'SumBackward', 'SigmoidBackward', 'BCEBackward']
+__all__ = ['Function', 'AddBackward', 'MulBackward', 'MatmulBackward', 'SumBackward', 'SigmoidBackward', 'BCEBackward',
+           'enable_autograd']
 
 # %% ../../modules/source/05_autograd/autograd_dev.ipynb 1
 import numpy as np
@@ -284,3 +299,198 @@ class BCEBackward(Function):
             
             return grad * grad_output,
         return None,
+
+# %% ../../modules/source/05_autograd/autograd_dev.ipynb 22
+def enable_autograd():
+    """
+    Enable gradient tracking for all Tensor operations.
+
+    This function enhances the existing Tensor class with autograd capabilities.
+    Call this once to activate gradients globally.
+
+    **What it does:**
+    - Replaces Tensor operations with gradient-tracking versions
+    - Adds backward() method for reverse-mode differentiation
+    - Enables computation graph building
+    - Maintains full backward compatibility
+
+    **After calling this:**
+    - Tensor operations will track computation graphs
+    - backward() method becomes available
+    - Gradients will flow through operations
+    - requires_grad=True enables tracking per tensor
+
+    **Example:**
+    ```python
+    enable_autograd()  # Call once
+    x = Tensor([2.0], requires_grad=True)
+    y = x * 3
+    y.backward()
+    print(x.grad)  # [3.0]
+    ```
+    """
+
+    # Check if already enabled
+    if hasattr(Tensor, '_autograd_enabled'):
+        print("⚠️ Autograd already enabled")
+        return
+
+    # Store original operations
+    _original_add = Tensor.__add__
+    _original_mul = Tensor.__mul__
+    _original_matmul = Tensor.matmul if hasattr(Tensor, 'matmul') else None
+
+    # Enhanced operations that track gradients
+    def tracked_add(self, other):
+        """
+        Addition with gradient tracking.
+        
+        Enhances the original __add__ method to build computation graphs
+        when requires_grad=True for any input.
+        """
+        # Convert scalar to Tensor if needed
+        if not isinstance(other, Tensor):
+            other = Tensor(other)
+
+        # Call original operation
+        result = _original_add(self, other)
+
+        # Track gradient if needed
+        if self.requires_grad or other.requires_grad:
+            result.requires_grad = True
+            result._grad_fn = AddBackward(self, other)
+
+        return result
+
+    def tracked_mul(self, other):
+        """
+        Multiplication with gradient tracking.
+        
+        Enhances the original __mul__ method to build computation graphs
+        when requires_grad=True for any input.
+        """
+        # Convert scalar to Tensor if needed for consistency
+        if not isinstance(other, Tensor):
+            other_tensor = Tensor(other)
+        else:
+            other_tensor = other
+
+        # Call original operation
+        result = _original_mul(self, other)
+
+        # Track gradient if needed
+        if self.requires_grad or (isinstance(other, Tensor) and other.requires_grad):
+            result.requires_grad = True
+            result._grad_fn = MulBackward(self, other)
+
+        return result
+
+    def tracked_matmul(self, other):
+        """
+        Matrix multiplication with gradient tracking.
+        
+        Enhances the original matmul method to build computation graphs
+        when requires_grad=True for any input.
+        """
+        if _original_matmul:
+            result = _original_matmul(self, other)
+        else:
+            # Fallback if matmul doesn't exist
+            result = Tensor(np.dot(self.data, other.data))
+
+        # Track gradient if needed
+        if self.requires_grad or other.requires_grad:
+            result.requires_grad = True
+            result._grad_fn = MatmulBackward(self, other)
+
+        return result
+
+    def sum_op(self, axis=None, keepdims=False):
+        """
+        Sum operation with gradient tracking.
+        
+        Creates a new sum method that builds computation graphs
+        when requires_grad=True.
+        """
+        result_data = np.sum(self.data, axis=axis, keepdims=keepdims)
+        result = Tensor(result_data)
+
+        if self.requires_grad:
+            result.requires_grad = True
+            result._grad_fn = SumBackward(self)
+
+        return result
+
+    def backward(self, gradient=None):
+        """
+        Compute gradients via backpropagation.
+
+        This is the key method that makes training possible!
+        It implements reverse-mode automatic differentiation.
+        
+        **Algorithm:**
+        1. Initialize gradient if not provided (for scalar outputs)
+        2. Accumulate gradient in self.grad
+        3. If this tensor has a _grad_fn, call it to propagate gradients
+        4. Recursively call backward() on parent tensors
+        
+        **Example:**
+        ```python
+        x = Tensor([2.0], requires_grad=True)
+        y = x * 3
+        y.backward()  # Computes gradients for x
+        print(x.grad)  # [3.0]
+        ```
+        """
+        # Only compute gradients if required
+        if not self.requires_grad:
+            return
+
+        # Initialize gradient if not provided (for scalar outputs)
+        if gradient is None:
+            if self.data.size == 1:
+                gradient = np.ones_like(self.data)
+            else:
+                raise ValueError("backward() requires gradient for non-scalar outputs")
+
+        # Initialize or accumulate gradient
+        if self.grad is None:
+            self.grad = np.zeros_like(self.data)
+        self.grad += gradient
+
+        # Propagate gradients through computation graph
+        if hasattr(self, '_grad_fn') and self._grad_fn:
+            grads = self._grad_fn.apply(gradient)
+
+            # Recursively call backward on parent tensors
+            for tensor, grad in zip(self._grad_fn.saved_tensors, grads):
+                if isinstance(tensor, Tensor) and tensor.requires_grad and grad is not None:
+                    tensor.backward(grad)
+
+    def zero_grad(self):
+        """
+        Reset gradients to zero.
+        
+        Call this before each backward pass to prevent gradient accumulation
+        from previous iterations.
+        """
+        self.grad = None
+
+    # Install enhanced operations
+    Tensor.__add__ = tracked_add
+    Tensor.__mul__ = tracked_mul
+    Tensor.matmul = tracked_matmul
+    Tensor.sum = sum_op
+    Tensor.backward = backward
+    Tensor.zero_grad = zero_grad
+
+    # Mark as enabled
+    Tensor._autograd_enabled = True
+
+    print("✅ Autograd enabled! Tensors now track gradients.")
+    print("   - Operations build computation graphs")
+    print("   - backward() computes gradients")
+    print("   - requires_grad=True enables tracking")
+
+# Auto-enable when module is imported
+enable_autograd()