diff --git a/.Jules/bolt.md b/.Jules/bolt.md
new file mode 100644
index 00000000..fbe4ec43
--- /dev/null
+++ b/.Jules/bolt.md
@@ -0,0 +1 @@
+## 2026-03-31 - Replace slow DataFrame .iterrows() loop with .apply() | Learning: Iterating through pandas DataFrames using .iterrows() is notoriously slow (essentially a Python loop over rows), whereas vectorization or .apply() is significantly faster and more idiomatic for pandas. | Action: Use .apply() with axis=1 for applying complex operations to DataFrame rows instead of explicit iteration.
diff --git a/monitoring/conversation_recommendation_optimizer.py b/monitoring/conversation_recommendation_optimizer.py
index 05cb405c..5c59e331 100644
--- a/monitoring/conversation_recommendation_optimizer.py
+++ b/monitoring/conversation_recommendation_optimizer.py
@@ -164,12 +164,10 @@ def _establish_quality_benchmarks(
         benchmarks = {}
 
         # Calculate quality scores for all conversations
-        quality_scores = []
-        for _, conv in conversations.iterrows():
-            score = self._calculate_conversation_quality_score(conv)
-            quality_scores.append(score)
-
-        conversations["quality_score"] = quality_scores
+        # ⚡ Bolt: Replaced slow .iterrows() loop with .apply() for significant performance gain
+        conversations["quality_score"] = conversations.apply(
+            self._calculate_conversation_quality_score, axis=1
+        )
 
         # Get top 10% as benchmark
         top_10_percent = conversations.nlargest(