diff --git a/backend/src/services.py b/backend/src/services.py
index 9ff5375..d9ae5fe 100644
--- a/backend/src/services.py
+++ b/backend/src/services.py
@@ -2,10 +2,16 @@
 import traceback
 import asyncio
 import time
+import logging
+import os
 from src.client import client
 from src.functions.functions import generate_code, run_locally, validate_output
 from src.workflows.workflow import AutonomousCodingWorkflow
 
+# Set up basic logging
+logging.basicConfig(level=logging.ERROR)
+logger = logging.getLogger(__name__)
+
 async def main():
     try:
         await client.start_service(
@@ -13,15 +19,28 @@ async def main():
             functions=[generate_code, run_locally, validate_output],
         )
     except Exception as e:
-        print(f"Error starting service: traceback: {traceback.format_exc()}")
-        print(f"Error starting service: {e}")
+        # Log the detailed traceback
+        logger.error("Error starting service:", exc_info=True)
+        
+        # In development environments, you might want to see the traceback
+        if os.environ.get('DEBUG') == 'true':
+            print(f"Error starting service: traceback: {traceback.format_exc()}")
+        else:
+            # In production, only show a generic message
+            print(f"Error starting service: {e}")
+        
         raise
 
 def run_services():
     try:
         asyncio.run(main())
     except Exception as e:
+        # Log the detailed error
+        logger.error("Service failed:", exc_info=True)
+        
+        # Simple error message for users
         print(f"Service failed: {e}")
+    
     # Keep the process alive for inspection
     while True:
         time.sleep(1)