diff --git a/.env.example b/.env.example
index f2cf73e5..981f1da2 100644
--- a/.env.example
+++ b/.env.example
@@ -33,4 +33,9 @@ EMBEDDING_DIMENSIONS=1536
 # - OPENAI_API_KEY (encrypted)
 # - MODEL_CHOICE 
 # - TRANSPORT settings
-# - RAG strategy flags (USE_CONTEXTUAL_EMBEDDINGS, USE_HYBRID_SEARCH, etc.)
\ No newline at end of file
+# - RAG strategy flags (USE_CONTEXTUAL_EMBEDDINGS, USE_HYBRID_SEARCH, etc.)
+# - Crawler settings:
+#   * CRAWL_MAX_CONCURRENT (default: 10) - Max concurrent pages per crawl operation
+#   * CRAWL_BATCH_SIZE (default: 50) - URLs processed per batch
+#   * MEMORY_THRESHOLD_PERCENT (default: 80) - Memory % before throttling
+#   * DISPATCHER_CHECK_INTERVAL (default: 0.5) - Memory check interval in seconds
\ No newline at end of file
diff --git a/python/src/server/services/crawling/strategies/batch.py b/python/src/server/services/crawling/strategies/batch.py
index a17124db..d97b0bc4 100644
--- a/python/src/server/services/crawling/strategies/batch.py
+++ b/python/src/server/services/crawling/strategies/batch.py
@@ -59,7 +59,7 @@ class BatchCrawlStrategy:
                 await progress_callback("error", 0, "Crawler not available")
             return []
 
-        # Load settings from database first
+        # Load settings from database - fail fast on configuration errors
         try:
             settings = await credential_service.get_credentials_by_category("rag_strategy")
             batch_size = int(settings.get("CRAWL_BATCH_SIZE", "50"))
@@ -67,11 +67,16 @@ class BatchCrawlStrategy:
                 max_concurrent = int(settings.get("CRAWL_MAX_CONCURRENT", "10"))
             memory_threshold = float(settings.get("MEMORY_THRESHOLD_PERCENT", "80"))
             check_interval = float(settings.get("DISPATCHER_CHECK_INTERVAL", "0.5"))
+        except (ValueError, KeyError, TypeError) as e:
+            # Critical configuration errors should fail fast in alpha
+            logger.error(f"Invalid crawl settings format: {e}", exc_info=True)
+            raise ValueError(f"Failed to load crawler configuration: {e}")
         except Exception as e:
-            logger.warning(f"Failed to load crawl settings: {e}, using defaults")
+            # For non-critical errors (e.g., network issues), use defaults but log prominently
+            logger.error(f"Failed to load crawl settings from database: {e}, using defaults", exc_info=True)
             batch_size = 50
             if max_concurrent is None:
-                max_concurrent = 10
+                max_concurrent = 10  # Safe default to prevent memory issues
             memory_threshold = 80.0
             check_interval = 0.5
             settings = {}  # Empty dict for defaults
diff --git a/python/src/server/services/crawling/strategies/recursive.py b/python/src/server/services/crawling/strategies/recursive.py
index 675c97f0..8b9cf93e 100644
--- a/python/src/server/services/crawling/strategies/recursive.py
+++ b/python/src/server/services/crawling/strategies/recursive.py
@@ -61,7 +61,7 @@ class RecursiveCrawlStrategy:
                 await progress_callback('error', 0, 'Crawler not available')
             return []
         
-        # Load settings from database
+        # Load settings from database - fail fast on configuration errors
         try:
             settings = await credential_service.get_credentials_by_category("rag_strategy")
             batch_size = int(settings.get("CRAWL_BATCH_SIZE", "50"))
@@ -69,11 +69,16 @@ class RecursiveCrawlStrategy:
                 max_concurrent = int(settings.get("CRAWL_MAX_CONCURRENT", "10"))
             memory_threshold = float(settings.get("MEMORY_THRESHOLD_PERCENT", "80"))
             check_interval = float(settings.get("DISPATCHER_CHECK_INTERVAL", "0.5"))
+        except (ValueError, KeyError, TypeError) as e:
+            # Critical configuration errors should fail fast in alpha
+            logger.error(f"Invalid crawl settings format: {e}", exc_info=True)
+            raise ValueError(f"Failed to load crawler configuration: {e}")
         except Exception as e:
-            logger.warning(f"Failed to load crawl settings: {e}, using defaults")
+            # For non-critical errors (e.g., network issues), use defaults but log prominently
+            logger.error(f"Failed to load crawl settings from database: {e}, using defaults", exc_info=True)
             batch_size = 50
             if max_concurrent is None:
-                max_concurrent = 10
+                max_concurrent = 10  # Safe default to prevent memory issues
             memory_threshold = 80.0
             check_interval = 0.5
             settings = {}  # Empty dict for defaults