diff --git a/src/google/adk/sessions/database_session_service.py b/src/google/adk/sessions/database_session_service.py
index c7c86e6e5f..6665669ed9 100644
--- a/src/google/adk/sessions/database_session_service.py
+++ b/src/google/adk/sessions/database_session_service.py
@@ -64,6 +64,7 @@
 _MARIADB_DIALECT = "mariadb"
 _MYSQL_DIALECT = "mysql"
 _POSTGRESQL_DIALECT = "postgresql"
+_DATABRICKS_DIALECT = "databricks"
 # Tuple key order for in-process per-session lock maps:
 # (app_name, user_id, session_id).
 _SessionLockKey: TypeAlias = tuple[str, str, str]
diff --git a/src/google/adk/sessions/schemas/shared.py b/src/google/adk/sessions/schemas/shared.py
index 25d4ea9e95..311c616ec6 100644
--- a/src/google/adk/sessions/schemas/shared.py
+++ b/src/google/adk/sessions/schemas/shared.py
@@ -37,6 +37,9 @@ def load_dialect_impl(self, dialect: Dialect):
     if dialect.name == "mysql":
       # Use LONGTEXT for MySQL to address the data too long issue
       return dialect.type_descriptor(mysql.LONGTEXT)
+    if dialect.name == "databricks":
+      # Databricks SQL stores JSON as STRING; use Text (the default)
+      return dialect.type_descriptor(Text)
     return dialect.type_descriptor(Text)  # Default to Text for other dialects
 
   def process_bind_param(self, value, dialect: Dialect):
@@ -64,4 +67,7 @@ class PreciseTimestamp(TypeDecorator):
   def load_dialect_impl(self, dialect):
     if dialect.name == "mysql":
       return dialect.type_descriptor(mysql.DATETIME(fsp=6))
+    if dialect.name == "databricks":
+      # Databricks TIMESTAMP type natively supports microsecond precision
+      return dialect.type_descriptor(DateTime)
     return self.impl