diff --git a/user_tools/src/spark_rapids_tools/tools/qualx/model.py b/user_tools/src/spark_rapids_tools/tools/qualx/model.py
index 9f9470a1f..4491545f2 100644
--- a/user_tools/src/spark_rapids_tools/tools/qualx/model.py
+++ b/user_tools/src/spark_rapids_tools/tools/qualx/model.py
@@ -347,7 +347,7 @@ def extract_model_features(
                 default_df = default_df.loc[~default_df.appName.str.startswith(f'{ds_name}:')]
             modified_default_df = default_split_fn(default_df)
             if modified_default_df.index.equals(default_df.index):
-                cpu_aug_tbl.update(default_df)
+                cpu_aug_tbl.update(modified_default_df)
                 cpu_aug_tbl.astype(df_schema)
             else:
                 raise ValueError('Default split_function unexpectedly modified row indices.')
diff --git a/user_tools/src/spark_rapids_tools/tools/qualx/preprocess.py b/user_tools/src/spark_rapids_tools/tools/qualx/preprocess.py
index d203bbc7f..320615b7c 100644
--- a/user_tools/src/spark_rapids_tools/tools/qualx/preprocess.py
+++ b/user_tools/src/spark_rapids_tools/tools/qualx/preprocess.py
@@ -30,7 +30,9 @@
     get_logger,
     get_dataset_platforms,
     load_plugin,
-    run_profiler_tool, log_fallback,
+    log_fallback,
+    run_profiler_tool,
+    RegexPattern
 )
 
 PREPROCESSED_FILE = 'preprocessed.parquet'
@@ -269,10 +271,12 @@ def infer_app_meta(eventlogs: List[str]) -> Mapping[str, Mapping]:
         app_meta_inner = {}
         for e in eventlog_list:
             parts = Path(e).parts
-            app_id_inner = parts[-1]
+            app_id_part = parts[-1]
+            match = RegexPattern.app_id.search(app_id_part)
+            app_id = match.group() if match else app_id_part
             run_type = parts[-2].upper()
             job_name = parts[-4]
-            app_meta_inner[app_id_inner] = {
+            app_meta_inner[app_id] = {
                 'jobName': job_name,
                 'runType': run_type,
                 'scaleFactor': 1,
diff --git a/user_tools/src/spark_rapids_tools/tools/qualx/util.py b/user_tools/src/spark_rapids_tools/tools/qualx/util.py
index 831721ff3..9bb8ee1a5 100644
--- a/user_tools/src/spark_rapids_tools/tools/qualx/util.py
+++ b/user_tools/src/spark_rapids_tools/tools/qualx/util.py
@@ -45,9 +45,9 @@ def get_logger(name: str) -> logging.Logger:
 
 @dataclass
 class RegexPattern:
-    app_id = re.compile(r'^app.*[_-][0-9]+[_-][0-9]+$')
-    profile = re.compile(r'^prof_[0-9]+_[0-9a-zA-Z]+$')
-    qual_tool = re.compile(r'^qual_[0-9]+_[0-9a-zA-Z]+$')
+    app_id = re.compile(r'app.*[_-][0-9]+[_-][0-9]+')
+    profile = re.compile(r'prof_[0-9]+_[0-9a-zA-Z]+')
+    qual_tool = re.compile(r'qual_[0-9]+_[0-9a-zA-Z]+')
     rapids_profile = re.compile(r'rapids_4_spark_profile')
     rapids_qual = re.compile(r'rapids_4_spark_qualification_output')
     qual_tool_metrics = re.compile(r'raw_metrics')