diff --git a/forum/backends/mysql/models.py b/forum/backends/mysql/models.py
index 806b4292..565d1de3 100644
--- a/forum/backends/mysql/models.py
+++ b/forum/backends/mysql/models.py
@@ -11,6 +11,7 @@
 from django.contrib.contenttypes.fields import GenericForeignKey, GenericRelation
 from django.contrib.contenttypes.models import ContentType
 from django.core.exceptions import ValidationError
+from django.db.models import F
 from django.db import models
 from django.db.models import QuerySet
 from django.utils import timezone
@@ -437,40 +438,33 @@ def get_sort_key(self) -> str:
     def get_list(**kwargs: Any) -> list[dict[str, Any]]:
         """
         Retrieves a list of all comments in the database based on provided filters.
-
-        Args:
-            kwargs: The filter arguments.
-
-        Returns:
-            A list of comments.
+        Optimized with select_related and DB-level sorting/pagination.
         """
+
         sort = kwargs.pop("sort", None)
         resp_skip = kwargs.pop("resp_skip", 0)
         resp_limit = kwargs.pop("resp_limit", None)
-        comments = Comment.objects.filter(**kwargs)
-        result = []
+
+        # select_related eliminates FK queries in to_dict()
+        comments = Comment.objects.filter(**kwargs).select_related(
+            "author", "parent", "comment_thread", "deleted_by"
+        )
+
+        # DB-level sorting instead of Python sorted()
         if sort:
             if sort == 1:
-                result = sorted(
-                    comments, key=lambda x: (x.sort_key is None, x.sort_key or "")
-                )
+                comments = comments.order_by(F("sort_key").asc(nulls_last=True))
             elif sort == -1:
-                result = sorted(
-                    comments,
-                    key=lambda x: (x.sort_key is None, x.sort_key or ""),
-                    reverse=True,
-                )
-
-        paginated_comments = result or list(comments)
+                comments = comments.order_by(F("sort_key").desc(nulls_last=True))
 
-        # Apply pagination if resp_limit is provided
+        # DB-level pagination instead of loading all then slicing
         if resp_limit is not None:
             resp_end = resp_skip + resp_limit
-            paginated_comments = result[resp_skip:resp_end]
-        elif resp_skip:  # If resp_limit is None but resp_skip is provided
-            paginated_comments = result[resp_skip:]
+            comments = comments[resp_skip:resp_end]
+        elif resp_skip:
+            comments = comments[resp_skip:]
 
-        return [content.to_dict() for content in paginated_comments]
+        return [content.to_dict() for content in comments]
 
     @staticmethod
     def get_list_total_count(**kwargs: Any) -> int:
diff --git a/forum/serializers/comment.py b/forum/serializers/comment.py
index 4c68b0f1..3fb4c8d4 100644
--- a/forum/serializers/comment.py
+++ b/forum/serializers/comment.py
@@ -86,21 +86,11 @@ def get_children(self, obj: Any) -> list[dict[str, Any]]:
         return list(serializer.data)
 
     def to_representation(self, instance: Any) -> dict[str, Any]:
+        """Return comment representation."""
         comment = super().to_representation(instance)
         comment.pop("historical_abuse_flaggers")
         if comment["parent_id"] == "None":
             comment["parent_id"] = None
-
-        thread = self.backend.get_thread(comment["thread_id"])
-        comment_from_db = self.backend.get_comment(comment["id"])
-        if (
-            not comment["endorsed"]
-            and comment_from_db
-            and "endorsement" not in comment_from_db
-            and thread
-            and thread["thread_type"] == "question"
-        ):
-            comment.pop("endorsement", None)
         return comment
 
     def create(self, validated_data: dict[str, Any]) -> Any: