diff --git a/.github/workflows/deploy.yml b/.github/workflows/deploy.yml
index c639b9d..e379b2d 100644
--- a/.github/workflows/deploy.yml
+++ b/.github/workflows/deploy.yml
@@ -31,10 +31,16 @@ jobs:
         with:
           python-version: '3.11'
           
-      - name: Install dependencies
+      - name: Setup Conda
+        uses: conda-incubator/setup-miniconda@v3
+        with:
+          auto-update-conda: true
+          python-version: '3.11'
+          
+      - name: Install book building dependencies
         run: |
-          python -m pip install --upgrade pip
-          pip install -r requirements.txt
+          conda env create -f requirements-book.yml
+          conda activate jupyter-book
           
       - name: Build the book
         run: |
diff --git a/README.md b/README.md
index 00af1c5..5b2d95e 100644
--- a/README.md
+++ b/README.md
@@ -76,4 +76,28 @@ Remember: Everyone learns differently. If this approach doesn't click with you,
 ### (...)
 
 ### 2. Here is how this course is structured and intended to be used. 
-This is important because if you go how it is intended you'll get the maximum amount of value out of this course.
\ No newline at end of file
+This is important because if you go how it is intended you'll get the maximum amount of value out of this course.
+
+## Installation 📦
+
+There are two ways to use this repository:
+
+1. **To run the code examples locally:**   ```bash
+   pip install -r requirements.txt   ```
+   This will install all the necessary ML/DL libraries needed to run the example code.
+
+2. **To build the book locally:**   ```bash
+   # Create and activate conda environment with book-building dependencies
+   conda env create -f requirements-book.yml
+   conda activate jupyter-book
+   
+   # Build the book
+   jupyter-book build . --all`
+
+The separation provides several benefits:
+1. Faster deployments since only necessary packages are installed
+2. Clearer separation of concerns between book building and code running
+3. Conda environment for book building helps avoid dependency conflicts
+4. Users can choose to install only what they need
+
+You might also want to add a `.gitignore` entry for the conda environment:
\ No newline at end of file
diff --git a/requirements-book.yml b/requirements-book.yml
new file mode 100644
index 0000000..e51b6ed
--- /dev/null
+++ b/requirements-book.yml
@@ -0,0 +1,14 @@
+name: jupyter-book
+channels:
+  - conda-forge
+dependencies:
+  - python=3.11
+  - jupyter-book>=0.13.0
+  - ghp-import
+  - sphinx>=4.0
+  - sphinx-book-theme
+  - myst-nb
+  - myst-parser
+  - nbformat
+  - nbconvert
+  - linkify-it-py 
\ No newline at end of file
diff --git a/requirements.txt b/requirements.txt
index 385a55e..7f7335b 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,16 +1,10 @@
-jupyter-book>=0.13.0
-ghp-import
-sphinx>=4.0
-sphinx-book-theme
-myst-nb
-myst-parser
-nbformat
-nbconvert
-linkify-it-py
-
-python==3.11.10
-matplotlib==3.9.2
+# Core ML/DL libraries
+torch==2.5.1
 numpy==2.1.3
 pandas==2.2.3
-requests==2.32.3
-torch==2.5.1
\ No newline at end of file
+
+# Visualization
+matplotlib==3.9.2
+
+# Other utilities
+requests==2.32.3
\ No newline at end of file