diff --git a/dataset.json b/dataset.json
new file mode 100644
index 0000000..0164533
--- /dev/null
+++ b/dataset.json
@@ -0,0 +1,20 @@
+{
+    "titles": [
+        "ranting about jira",
+        "siting in a meeting",
+        "squashing bugs",
+        "centering divs",
+        "waiting for staging deployment",
+        "writing tests",
+        "pushing directly to prod",
+        "doing code review",
+        "deleting jira tickets",
+        "Creating a pull request",
+        "forgot to run DB migrations",
+        "Broke production",
+        "running on staging ENV",
+        "merging git branches",
+        "releasing the MVP",
+        "finding JIRA tickets"
+    ]
+}
\ No newline at end of file
diff --git a/main.py b/main.py
index ae9b4c4..9b04b1b 100644
--- a/main.py
+++ b/main.py
@@ -2,35 +2,20 @@ import praw
 import os
 import random
 import time
-
-titles = [
-    "ranting about jira",
-    "siting in a meeting",
-    "squashing bugs",
-    "centering divs",
-    "waiting for staging deployment",
-    "writing tests",
-    "pushing directly to prod",
-    "doing code review",
-    "deleting jira tickets",
-    "Creating a pull request",
-    "forgot to run DB migrations",
-    "Broke production",
-    "running on staging ENV",
-    "merging git branches",
-    "releasing the MVP",
-    "finding JIRA tickets"
-]
+import json
 
 client_id = os.environ["REDDIT_CLIENT_ID"]
 client_secret = os.environ["REDDIT_CLIENT_SECRET"]
 reddit_pass = os.environ["REDDIT_PASSWORD"]
 
 def get_titles():
+    with open('dataset.json', 'r') as f:
+         data = json.load(f)
+
+    titles = data['titles']
     currentlyViewingText, subscribersText = random.sample(titles, 2)
     return [currentlyViewingText, subscribersText]
 
-
 def update_titles():
     reddit = praw.Reddit(
         client_id=client_id,
@@ -53,4 +38,5 @@ def update_titles():
     widgets.refresh()
     widgets.id_card.mod.update(subscribersText=titles[1])
 
-update_titles()
\ No newline at end of file
+
+update_titles()