diff --git a/.gitignore b/.gitignore
index dc55edb..5ad18c9 100644
--- a/.gitignore
+++ b/.gitignore
@@ -51,6 +51,7 @@ htmlcov/
 # IDEs
 .vscode/
 .idea/
+*.ipynb
 
 # OS
 .DS_Store
@@ -64,3 +65,6 @@ data/embedding_dataset_from_questions_all_chunks_no_unclassified.json
 # Git
 .specify
 .github/prompts
+
+# Logs
+*.log
\ No newline at end of file
diff --git a/data/.gitignore b/data/.gitignore
index 9466895..5918f22 100644
--- a/data/.gitignore
+++ b/data/.gitignore
@@ -29,3 +29,7 @@
 /prd_documents
 /prd_dedup
 /annotations
+/processed-james
+/processed-sgp
+/raw-prd-sgp
+/esim_prd_synthetic_qa_with_subgroups
diff --git a/data/raw-prd-sgp.dvc b/data/raw-prd-sgp.dvc
new file mode 100644
index 0000000..5f3045d
--- /dev/null
+++ b/data/raw-prd-sgp.dvc
@@ -0,0 +1,6 @@
+outs:
+- md5: cac377080c3ef911d358961d6bfd19ac.dir
+  size: 32930661
+  nfiles: 29
+  hash: md5
+  path: raw-prd-sgp
diff --git a/data/working_groups_mapping.json.dvc b/data/working_groups_mapping.json.dvc
index 8a5b4d9..cd3f2f9 100644
--- a/data/working_groups_mapping.json.dvc
+++ b/data/working_groups_mapping.json.dvc
@@ -1,5 +1,5 @@
 outs:
-- md5: b70592bdeac5c03634a60d09d0a8fbc7
-  size: 9901
+- md5: 43d32045e3db97435255cb18015a4e1f
+  size: 10411
   hash: md5
   path: working_groups_mapping.json
diff --git a/gsma_dataset_creation/cli.py b/gsma_dataset_creation/cli.py
index 088937d..09110cf 100644
--- a/gsma_dataset_creation/cli.py
+++ b/gsma_dataset_creation/cli.py
@@ -61,9 +61,29 @@ def setup_logging(log_level: str) -> None:
     logger.add(
         lambda msg: typer.echo(msg, err=True),
         level=log_level.upper(),
-        format="<green>{time:YYYY-MM-DD HH:mm:ss}</green> | <level>{level: <8}</level> | {message}",
+        format = (
+            "<green>{time:YYYY-MM-DD HH:mm:ss}</green> | "
+            "<level>{level:<8}</level> | "
+            "<cyan>{file}:{line}</cyan> | "
+            "{message}"
+        )
     )
 
+    # File output
+    logger.add(
+        "GSMA-data-creation.log",
+        level=log_level.upper(),
+        format=(
+            "<green>{time:YYYY-MM-DD HH:mm:ss}</green> | "
+            "<level>{level:<8}</level> | "
+            "<cyan>{file}:{line}</cyan> | "
+            "{message}"
+        ),
+        rotation="20 MB",
+        retention="30 days",
+        compression="zip",
+        enqueue=True  # Safe for async/multiprocessing
+    )
 
 @app.command()
 def process(
diff --git a/gsma_dataset_creation/qa_schema.py b/gsma_dataset_creation/qa_schema.py
index 05ff3ca..202fa4f 100644
--- a/gsma_dataset_creation/qa_schema.py
+++ b/gsma_dataset_creation/qa_schema.py
@@ -37,10 +37,11 @@ def get_qa_response_schema(config: QAConfig) -> Dict[str, Any]:
                         "question_type": {
                             "type": "string",
                             "enum": [
-                                "factual",
-                                "comprehension",
-                                "analytical",
-                                "definition",
+                                "multiple_choice",
+                                "multiple_choice_explained",
+                                "true_false",
+                                "open_ended",
+                                "open_ended_no_context"
                             ],
                         },
                     },
diff --git a/gsma_dataset_creation/question_generator.py b/gsma_dataset_creation/question_generator.py
index 005c14d..1913248 100644
--- a/gsma_dataset_creation/question_generator.py
+++ b/gsma_dataset_creation/question_generator.py
@@ -12,6 +12,7 @@
 from datetime import datetime
 from pathlib import Path
 from typing import Any, Dict, List, Optional, cast
+import traceback
 
 from loguru import logger
 from openai import (
@@ -148,6 +149,7 @@ async def _api_call() -> Any:
             )
         except APIStatusError as e:
             # Check if it's a structured output compatibility issue
+            logger.error(traceback.format_exc())
             error_msg = str(e.message) if hasattr(e, "message") else str(e)
             if e.status_code == 400 and any(
                 keyword in error_msg.lower()
@@ -169,6 +171,7 @@ async def _api_call() -> Any:
             # Re-raise other API status errors
             raise
         except Exception as e:
+            logger.error(traceback.format_exc())
             # For unexpected exceptions, log details for debugging
             logger.debug(
                 f"Unexpected exception type: {type(e).__name__}, message: {str(e)}"
@@ -194,6 +197,7 @@ async def _api_call() -> Any:
                         **extra_params,
                     )
                 except Exception as fallback_error:
+                    logger.error(traceback.format_exc())
                     logger.error(f"Fallback also failed: {fallback_error}")
                     raise fallback_error
             raise
diff --git a/gsma_dataset_creation/retry_utils.py b/gsma_dataset_creation/retry_utils.py
index af75df7..9a8e498 100644
--- a/gsma_dataset_creation/retry_utils.py
+++ b/gsma_dataset_creation/retry_utils.py
@@ -129,6 +129,8 @@ async def wrapper(*args: Any, **kwargs: Any) -> Any:
                     f"Failed {func.__name__} after {config.max_retries} retries: "
                     f"{type(e).__name__}: {e}"
                 )
+                import traceback
+                logger.error(traceback.format_exc())
                 raise
 
         return wrapper
diff --git a/gsma_dataset_creation/similarity/similarity_calculator.py b/gsma_dataset_creation/similarity/similarity_calculator.py
index 23aa3d0..5d2ec57 100644
--- a/gsma_dataset_creation/similarity/similarity_calculator.py
+++ b/gsma_dataset_creation/similarity/similarity_calculator.py
@@ -5,6 +5,7 @@
 using efficient approximate nearest neighbor search.
 """
 
+import os
 import json
 from datetime import datetime
 from pathlib import Path
@@ -97,6 +98,16 @@ def build_faiss_index(
         norms = np.linalg.norm(embeddings, axis=1, keepdims=True)
         embeddings = embeddings / norms
 
+        # Minimum number of points needed per centroid for effective clustering
+        min_points_per_centroid = 39
+
+        # Adjust index type if dataset too small
+        if n_chunks < min_points_per_centroid:
+            logger.debug(
+                f"Dataset size {n_chunks} is sufficient for IVFFlat index"
+            )
+            self.faiss_index_type = "Flat"
+    
         # Build index based on type
         if self.faiss_index_type == "Flat":
             # Flat index for exact search (slower but accurate)
@@ -112,6 +123,10 @@ def build_faiss_index(
                 if self.nlist > n_chunks // 10:
                     self.nlist = max(1, n_chunks // 10)
 
+            # Ensure nlist is appropriate for dataset size
+            if self.nlist * min_points_per_centroid > n_chunks:
+                self.nlist = max(1, n_chunks // min_points_per_centroid)
+
             logger.info(f"Building IVFFlat index with nlist={self.nlist}")
 
             # Create quantizer and index
@@ -181,8 +196,15 @@ def compute_similarities(
         k_search = min(
             self.k + 1, len(self.chunk_metadata)
         )  # +1 to exclude self-similarity
+        
+        # In some environment, or when Faiss is built with MKL, 
+        # this can cause segmentation faults, crashes or weird failures during index.search().
+        # This change limits OpenMP for multithreading to a single thread
+        os.environ.setdefault("OMP_NUM_THREADS", "1")
+
         scores, indices = self.faiss_index.search(query_embeddings, k_search)
 
+
         # Process results
         all_similarities = []
         similarities_computed = 0
diff --git a/gsma_dataset_creation/similarity_cli.py b/gsma_dataset_creation/similarity_cli.py
index 0bc232e..3e41ee2 100644
--- a/gsma_dataset_creation/similarity_cli.py
+++ b/gsma_dataset_creation/similarity_cli.py
@@ -17,7 +17,22 @@ def setup_logging(log_level: str) -> None:
     logger.add(
         lambda msg: typer.echo(msg, err=True),
         level=log_level.upper(),
-        format="<green>{time:YYYY-MM-DD HH:mm:ss}</green> | <level>{level: <8}</level> | {message}",
+        format="<green>{time:YYYY-MM-DD HH:mm:ss}</green> | <level>{level: <8}</level> | <cyan>{file}:{line}</cyan> | {message}",
+    )
+    # File output
+    logger.add(
+        "GSMA-data-creation.log",
+        level=log_level.upper(),
+        format=(
+            "<green>{time:YYYY-MM-DD HH:mm:ss}</green> | "
+            "<level>{level:<8}</level> | "
+            "<cyan>{file}:{line}</cyan> | "
+            "{message}"
+        ),
+        rotation="20 MB",
+        retention="30 days",
+        compression="zip",
+        enqueue=True  # Safe for async/multiprocessing
     )
 
 
@@ -534,6 +549,8 @@ def ranker(
         raise typer.Exit(1)
     except Exception as e:
         logger.error(f"❌ Unexpected error: {e}")
+        import traceback
+        logger.error(traceback.format_exc())
         raise typer.Exit(1)
 
 
diff --git a/pipelines/annotation/dvc.lock b/pipelines/annotation/dvc.lock
index 7330482..e205328 100644
--- a/pipelines/annotation/dvc.lock
+++ b/pipelines/annotation/dvc.lock
@@ -1,32 +1,33 @@
 schema: '2.0'
 stages:
   add_subgroups:
-    cmd: uv run gsma add-subgroup-to-dataset --dataset-repo mantisnlp/gsma_prd_synthetic_qa
-      --working-groups data/working_groups_mapping.json --output data/gsma_prd_synthetic_qa_with_subgroups
-      --log-level INFO
+    cmd: uv run gsma add-subgroup-to-dataset --dataset-repo 
+      jamesaidev001/esim_prd_synthetic_qa --working-groups 
+      data/working_groups_mapping.json --output 
+      data/esim_prd_synthetic_qa_with_subgroups --log-level INFO
     deps:
     - path: data/working_groups_mapping.json
       hash: md5
-      md5: 2adbc799e706c013db55d1fb3338878c
-      size: 10629
+      md5: 43d32045e3db97435255cb18015a4e1f
+      size: 10411
     - path: gsma_dataset_creation/cli.py
       hash: md5
-      md5: 1bfad6c8fd2ce3455f7cdf9a1f1e0518
-      size: 80990
+      md5: e1d291f8c6f59a65f0a0914c283a3968
+      size: 82670
     - path: gsma_dataset_creation/subgroup_adder.py
       hash: md5
       md5: 6d25cfda36dd581cf975bd4453a850eb
       size: 14752
     outs:
-    - path: data/gsma_prd_synthetic_qa_with_subgroups
+    - path: data/esim_prd_synthetic_qa_with_subgroups
       hash: md5
-      md5: b1d1eff79e4d53a1cdcab707c068f5fe.dir
-      size: 423824806
+      md5: e4df70e8c255276e52c4e2af1347f112.dir
+      size: 21318642
       nfiles: 4
   upload_tsg_annotation:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_with_subgroups
-      --subgroup TSG --sample-size 100 --dataset-name-prefix gsma_annotation_tsg --logger-level
-      INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_with_subgroups --subgroup TSG --sample-size 100 
+      --dataset-name-prefix gsma_annotation_tsg --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_with_subgroups
       hash: md5
@@ -42,9 +43,9 @@ stages:
       md5: 3afc719e94c38002e8a124fe99344648
       size: 13916
   upload_annotation@@FASG:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup FASG --sample-size 1000 --dataset-name-prefix gsma_annotation_FASG
-      --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup FASG --sample-size 
+      1000 --dataset-name-prefix gsma_annotation_FASG --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -60,9 +61,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@PQTN:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup PQTN --sample-size 1000 --dataset-name-prefix gsma_annotation_PQTN
-      --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup PQTN --sample-size 
+      1000 --dataset-name-prefix gsma_annotation_PQTN --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -78,9 +79,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@TSG:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup TSG --sample-size 1000 --dataset-name-prefix gsma_annotation_TSG
-      --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup TSG --sample-size 
+      1000 --dataset-name-prefix gsma_annotation_TSG --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -96,9 +97,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@SAM:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup SAM --sample-size 1000 --dataset-name-prefix gsma_annotation_SAM
-      --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup SAM --sample-size 
+      1000 --dataset-name-prefix gsma_annotation_SAM --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -114,9 +115,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@eSim:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_with_subgroups
-      --subgroup eSim --sample-size 1000 --dataset-name-prefix gsma_annotation_eSim
-      --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_with_subgroups --subgroup eSim --sample-size 1000 
+      --dataset-name-prefix gsma_annotation_eSim --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_with_subgroups
       hash: md5
@@ -132,9 +133,9 @@ stages:
       md5: b0e96df363386ac9af0328636a6c3e3b
       size: 20248
   upload_annotation@@IG:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup IG --sample-size 1000 --dataset-name-prefix gsma_annotation_IG --logger-level
-      INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup IG --sample-size 1000
+      --dataset-name-prefix gsma_annotation_IG --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -150,9 +151,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@SG:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup SG --sample-size 1000 --dataset-name-prefix gsma_annotation_SG --logger-level
-      INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup SG --sample-size 1000
+      --dataset-name-prefix gsma_annotation_SG --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -168,9 +169,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@NG:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup NG --sample-size 1000 --dataset-name-prefix gsma_annotation_NG --logger-level
-      INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup NG --sample-size 1000
+      --dataset-name-prefix gsma_annotation_NG --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -186,9 +187,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@IDS:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup IDS --sample-size 1000 --dataset-name-prefix gsma_annotation_IDS
-      --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup IDS --sample-size 
+      1000 --dataset-name-prefix gsma_annotation_IDS --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -204,9 +205,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@RCS:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup RCS --sample-size 1000 --dataset-name-prefix gsma_annotation_RCS
-      --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup RCS --sample-size 
+      1000 --dataset-name-prefix gsma_annotation_RCS --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -222,9 +223,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@OPG:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup OPG --sample-size 1000 --dataset-name-prefix gsma_annotation_OPG
-      --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup OPG --sample-size 
+      1000 --dataset-name-prefix gsma_annotation_OPG --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -240,9 +241,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@WAS:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup WAS --sample-size 1000 --dataset-name-prefix gsma_annotation_WAS
-      --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup WAS --sample-size 
+      1000 --dataset-name-prefix gsma_annotation_WAS --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -258,9 +259,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@SAS:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup SAS --sample-size 1000 --dataset-name-prefix gsma_annotation_SAS
-      --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup SAS --sample-size 
+      1000 --dataset-name-prefix gsma_annotation_SAS --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -276,9 +277,10 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@CERTIFICATION_POLICY:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup CERTIFICATION_POLICY --sample-size 1000 --dataset-name-prefix gsma_annotation_CERTIFICATION_POLICY
-      --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup CERTIFICATION_POLICY 
+      --sample-size 1000 --dataset-name-prefix 
+      gsma_annotation_CERTIFICATION_POLICY --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -294,9 +296,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@EID:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup EID --sample-size 1000 --dataset-name-prefix gsma_annotation_EID
-      --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup EID --sample-size 
+      1000 --dataset-name-prefix gsma_annotation_EID --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -312,8 +314,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@COMPLIANCE:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup COMPLIANCE --sample-size 1000 --dataset-name-prefix gsma_annotation_COMPLIANCE
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup COMPLIANCE 
+      --sample-size 1000 --dataset-name-prefix gsma_annotation_COMPLIANCE 
       --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -330,8 +333,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@REQUIREMENTS:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup REQUIREMENTS --sample-size 1000 --dataset-name-prefix gsma_annotation_REQUIREMENTS
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup REQUIREMENTS 
+      --sample-size 1000 --dataset-name-prefix gsma_annotation_REQUIREMENTS 
       --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -348,9 +352,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@ESA:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup ESA --sample-size 1000 --dataset-name-prefix gsma_annotation_ESA
-      --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup ESA --sample-size 
+      1000 --dataset-name-prefix gsma_annotation_ESA --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -366,8 +370,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@TECHNICAL:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup TECHNICAL --sample-size 1000 --dataset-name-prefix gsma_annotation_TECHNICAL
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup TECHNICAL 
+      --sample-size 1000 --dataset-name-prefix gsma_annotation_TECHNICAL 
       --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -384,9 +389,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation@@TESTING:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup TESTING --sample-size 1000 --dataset-name-prefix gsma_annotation_TESTING
-      --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup TESTING --sample-size
+      1000 --dataset-name-prefix gsma_annotation_TESTING --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -402,9 +407,9 @@ stages:
       md5: f116942d041709137200bf5bd1b4fec8
       size: 18809
   upload_annotation_working_groups@@IG:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup IG --sample-size 1000 --dataset-name-prefix gsma_annotation_IG --logger-level
-      INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup IG --sample-size 1000
+      --dataset-name-prefix gsma_annotation_IG --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -420,8 +425,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_ng_subgroups@@Oig_isig:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Oig_isig --working-group NG --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Oig_isig 
+      --working-group NG --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_ng_Oig_isig --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -438,9 +444,10 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_ng_subgroups@@Other:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Other --working-group NG --sample-size 1000 --dataset-name-prefix
-      gsma_annotation_ng_Other --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Other --working-group
+      NG --sample-size 1000 --dataset-name-prefix gsma_annotation_ng_Other 
+      --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -456,9 +463,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_working_groups@@SG:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup SG --sample-size 1000 --dataset-name-prefix gsma_annotation_SG --logger-level
-      INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup SG --sample-size 1000
+      --dataset-name-prefix gsma_annotation_SG --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -474,8 +481,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_esim_subgroups@@Security:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Security --working-group eSIM --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Security 
+      --working-group eSIM --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_esim_Security --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -492,9 +500,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_working_groups@@SAS:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup SAS --sample-size 1000 --dataset-name-prefix gsma_annotation_SAS
-      --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup SAS --sample-size 
+      1000 --dataset-name-prefix gsma_annotation_SAS --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -510,8 +518,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_esim_subgroups@@Esa_scheme:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Esa_scheme --working-group eSIM --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Esa_scheme 
+      --working-group eSIM --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_esim_Esa_scheme --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -528,8 +537,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_esim_subgroups@@Certification_policy:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Certification_policy --working-group eSIM --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Certification_policy 
+      --working-group eSIM --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_esim_Certification_policy --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -546,8 +556,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_ng_subgroups@@Upg_isig:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Upg_isig --working-group NG --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Upg_isig 
+      --working-group NG --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_ng_Upg_isig --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -564,8 +575,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_esim_subgroups@@Requirements:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Requirements --working-group eSIM --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Requirements 
+      --working-group eSIM --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_esim_Requirements --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -582,8 +594,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_esim_subgroups@@Testing:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Testing --working-group eSIM --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Testing 
+      --working-group eSIM --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_esim_Testing --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -600,8 +613,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_ng_subgroups@@Mc_isig:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Mc_isig --working-group NG --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Mc_isig 
+      --working-group NG --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_ng_Mc_isig --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -618,8 +632,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_ng_subgroups@@Tg:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Tg --working-group NG --sample-size 1000 --dataset-name-prefix gsma_annotation_ng_Tg
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Tg --working-group NG
+      --sample-size 1000 --dataset-name-prefix gsma_annotation_ng_Tg 
       --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -636,8 +651,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_esim_subgroups@@Compliance:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Compliance --working-group eSIM --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Compliance 
+      --working-group eSIM --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_esim_Compliance --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -654,8 +670,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_ng_subgroups@@Nrg_isig:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Nrg_isig --working-group NG --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Nrg_isig 
+      --working-group NG --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_ng_Nrg_isig --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -672,9 +689,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_working_groups@@PQTN:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup PQTN --sample-size 1000 --dataset-name-prefix gsma_annotation_PQTN
-      --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup PQTN --sample-size 
+      1000 --dataset-name-prefix gsma_annotation_PQTN --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -690,26 +707,28 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_esim_subgroups@Compliance:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Compliance --working-group eSIM --sample-size 1000 --dataset-name-prefix
-      gsma_annotation_esim_Compliance --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/esim_prd_synthetic_qa_with_subgroups --subgroup Compliance 
+      --working-group eSim --sample-size 1000 --dataset-name-prefix 
+      esim_Compliance --logger-level DEBUG
     deps:
-    - path: data/gsma_prd_synthetic_qa_with_subgroups
+    - path: data/esim_prd_synthetic_qa_with_subgroups
       hash: md5
-      md5: a683e4a7744b9c6fbf18c5b4eceba3dc.dir
-      size: 423827710
-      nfiles: 5
+      md5: 90bab7cea26f10c19bec4eeb4e7cd8ca.dir
+      size: 21358490
+      nfiles: 9
     - path: gsma_dataset_creation/argilla_cli.py
       hash: md5
-      md5: 77a9aaa9df4253afa88578dc676b4e27
-      size: 17579
+      md5: cf30c259ec2abc16adad41b33830495b
+      size: 57170
     - path: gsma_dataset_creation/validation/argilla_subgroup_uploader.py
       hash: md5
-      md5: ca99922bcb3c06c428edc22d8e8ec7ad
-      size: 20868
+      md5: 630c3037409fc170eeb6ba05357087aa
+      size: 22164
   upload_annotation_esim_subgroups@Esa_scheme:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Esa_scheme --working-group eSIM --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Esa_scheme 
+      --working-group eSIM --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_esim_Esa_scheme --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -726,44 +745,47 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_esim_subgroups@Technical:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Technical --working-group eSIM --sample-size 1000 --dataset-name-prefix
-      gsma_annotation_esim_Technical --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/esim_prd_synthetic_qa_with_subgroups --subgroup Technical 
+      --working-group eSim --sample-size 1000 --dataset-name-prefix 
+      esim_Technical --logger-level DEBUG
     deps:
-    - path: data/gsma_prd_synthetic_qa_with_subgroups
+    - path: data/esim_prd_synthetic_qa_with_subgroups
       hash: md5
-      md5: 021a45de31d599951e8c55d21411cdd9.dir
-      size: 423876334
+      md5: b0d1d9b50e059bbdedcf50dc972a095e.dir
+      size: 21350978
       nfiles: 7
     - path: gsma_dataset_creation/argilla_cli.py
       hash: md5
-      md5: 77a9aaa9df4253afa88578dc676b4e27
-      size: 17579
+      md5: cf30c259ec2abc16adad41b33830495b
+      size: 57170
     - path: gsma_dataset_creation/validation/argilla_subgroup_uploader.py
       hash: md5
-      md5: ca99922bcb3c06c428edc22d8e8ec7ad
-      size: 20868
+      md5: 630c3037409fc170eeb6ba05357087aa
+      size: 22164
   upload_annotation_esim_subgroups@Testing:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Testing --working-group eSIM --sample-size 1000 --dataset-name-prefix
-      gsma_annotation_esim_Testing --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/esim_prd_synthetic_qa_with_subgroups --subgroup Testing 
+      --working-group eSim --sample-size 1000 --dataset-name-prefix esim_Testing
+      --logger-level DEBUG
     deps:
-    - path: data/gsma_prd_synthetic_qa_with_subgroups
+    - path: data/esim_prd_synthetic_qa_with_subgroups
       hash: md5
-      md5: 571a5ce098aad2fc70df866491e3aa38.dir
-      size: 423892630
+      md5: bda3d5917c494486079b70900772e88b.dir
+      size: 21357146
       nfiles: 8
     - path: gsma_dataset_creation/argilla_cli.py
       hash: md5
-      md5: 77a9aaa9df4253afa88578dc676b4e27
-      size: 17579
+      md5: cf30c259ec2abc16adad41b33830495b
+      size: 57170
     - path: gsma_dataset_creation/validation/argilla_subgroup_uploader.py
       hash: md5
-      md5: ca99922bcb3c06c428edc22d8e8ec7ad
-      size: 20868
+      md5: 630c3037409fc170eeb6ba05357087aa
+      size: 22164
   upload_annotation_esim_subgroups@Eid_scheme:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Eid_scheme --working-group eSIM --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Eid_scheme 
+      --working-group eSIM --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_esim_Eid_scheme --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -780,26 +802,28 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_esim_subgroups@Requirements:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Requirements --working-group eSIM --sample-size 1000 --dataset-name-prefix
-      gsma_annotation_esim_Requirements --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/esim_prd_synthetic_qa_with_subgroups --subgroup Requirements 
+      --working-group eSim --sample-size 1000 --dataset-name-prefix 
+      esim_Requirements --logger-level DEBUG
     deps:
-    - path: data/gsma_prd_synthetic_qa_with_subgroups
+    - path: data/esim_prd_synthetic_qa_with_subgroups
       hash: md5
-      md5: 15cf4b5c8b536bc22d68a5632f036423.dir
-      size: 423909206
-      nfiles: 10
+      md5: 5753034ec4f32916361da8666a21e06b.dir
+      size: 21326074
+      nfiles: 6
     - path: gsma_dataset_creation/argilla_cli.py
       hash: md5
-      md5: 77a9aaa9df4253afa88578dc676b4e27
-      size: 17579
+      md5: cf30c259ec2abc16adad41b33830495b
+      size: 57170
     - path: gsma_dataset_creation/validation/argilla_subgroup_uploader.py
       hash: md5
-      md5: ca99922bcb3c06c428edc22d8e8ec7ad
-      size: 20868
+      md5: 630c3037409fc170eeb6ba05357087aa
+      size: 22164
   upload_annotation_esim_subgroups@Certification_policy:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Certification_policy --working-group eSIM --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Certification_policy 
+      --working-group eSIM --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_esim_Certification_policy --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -816,27 +840,29 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_esim_subgroups@Security:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Security --working-group eSIM --sample-size 1000 --dataset-name-prefix
-      gsma_annotation_esim_Security --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/esim_prd_synthetic_qa_with_subgroups --subgroup Security 
+      --working-group eSim --sample-size 1000 --dataset-name-prefix 
+      esim_Security --logger-level DEBUG
     deps:
-    - path: data/gsma_prd_synthetic_qa_with_subgroups
+    - path: data/esim_prd_synthetic_qa_with_subgroups
       hash: md5
-      md5: 1ca6a5b2083300b20e253a9050aaedcc.dir
-      size: 423920998
-      nfiles: 12
+      md5: 19c5a9baedceb8e4debde5b306d03487.dir
+      size: 21324010
+      nfiles: 5
     - path: gsma_dataset_creation/argilla_cli.py
       hash: md5
-      md5: 77a9aaa9df4253afa88578dc676b4e27
-      size: 17579
+      md5: cf30c259ec2abc16adad41b33830495b
+      size: 57170
     - path: gsma_dataset_creation/validation/argilla_subgroup_uploader.py
       hash: md5
-      md5: ca99922bcb3c06c428edc22d8e8ec7ad
-      size: 20868
+      md5: 630c3037409fc170eeb6ba05357087aa
+      size: 22164
   upload_annotation_ng_subgroups@Isag:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Isag --working-group NG --sample-size 1000 --dataset-name-prefix
-      gsma_annotation_ng_Isag --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Isag --working-group 
+      NG --sample-size 1000 --dataset-name-prefix gsma_annotation_ng_Isag 
+      --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -852,9 +878,10 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_ng_subgroups@Other:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Other --working-group NG --sample-size 1000 --dataset-name-prefix
-      gsma_annotation_ng_Other --logger-level INFO
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Other --working-group
+      NG --sample-size 1000 --dataset-name-prefix gsma_annotation_ng_Other 
+      --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
       hash: md5
@@ -870,8 +897,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_ng_subgroups@Tg:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Tg --working-group NG --sample-size 1000 --dataset-name-prefix gsma_annotation_ng_Tg
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Tg --working-group NG
+      --sample-size 1000 --dataset-name-prefix gsma_annotation_ng_Tg 
       --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -888,8 +916,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_ng_subgroups@Oig_isig:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Oig_isig --working-group NG --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Oig_isig 
+      --working-group NG --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_ng_Oig_isig --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -906,8 +935,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_ng_subgroups@Nrg_isig:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Nrg_isig --working-group NG --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Nrg_isig 
+      --working-group NG --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_ng_Nrg_isig --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -924,8 +954,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_ng_subgroups@Upg_isig:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Upg_isig --working-group NG --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Upg_isig 
+      --working-group NG --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_ng_Upg_isig --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -942,8 +973,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_ng_subgroups@Mc_isig:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Mc_isig --working-group NG --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Mc_isig 
+      --working-group NG --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_ng_Mc_isig --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -960,8 +992,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_ng_subgroups@Ursp_isig:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Ursp_isig --working-group NG --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Ursp_isig 
+      --working-group NG --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_ng_Ursp_isig --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -978,8 +1011,9 @@ stages:
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
   upload_annotation_ng_subgroups@Fnr_isig:
-    cmd: uv run gsma argilla upload-by-subgroup --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
-      --subgroup Fnr_isig --working-group NG --sample-size 1000 --dataset-name-prefix
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/gsma_prd_synthetic_qa_with_subgroups --subgroup Fnr_isig 
+      --working-group NG --sample-size 1000 --dataset-name-prefix 
       gsma_annotation_ng_Fnr_isig --logger-level INFO
     deps:
     - path: data/gsma_prd_synthetic_qa_with_subgroups
@@ -995,3 +1029,98 @@ stages:
       hash: md5
       md5: ca99922bcb3c06c428edc22d8e8ec7ad
       size: 20868
+  upload_annotation_esim_subgroups@Certification:
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/esim_prd_synthetic_qa_with_subgroups --subgroup Certification 
+      --working-group eSim --sample-size 1000 --dataset-name-prefix 
+      esim_Certification --logger-level DEBUG
+    deps:
+    - path: data/esim_prd_synthetic_qa_with_subgroups
+      hash: md5
+      md5: 0b6db7608974cd2d0444a389e326bd97.dir
+      size: 21361810
+      nfiles: 12
+    - path: gsma_dataset_creation/argilla_cli.py
+      hash: md5
+      md5: 96da29c6bf660ee8add034e08fec92ab
+      size: 58419
+    - path: gsma_dataset_creation/validation/argilla_subgroup_uploader.py
+      hash: md5
+      md5: 71199696b1c81567d1da0a9da07391b1
+      size: 22119
+  upload_annotation_esim_subgroups@EID_Scheme:
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/esim_prd_synthetic_qa_with_subgroups --subgroup EID_Scheme 
+      --working-group eSim --sample-size 1000 --dataset-name-prefix 
+      esim_EID_Scheme --logger-level DEBUG
+    deps:
+    - path: data/esim_prd_synthetic_qa_with_subgroups
+      hash: md5
+      md5: 9888acfb189bda92e3a61dd70ca20af6.dir
+      size: 21358794
+      nfiles: 10
+    - path: gsma_dataset_creation/argilla_cli.py
+      hash: md5
+      md5: cf30c259ec2abc16adad41b33830495b
+      size: 57170
+    - path: gsma_dataset_creation/validation/argilla_subgroup_uploader.py
+      hash: md5
+      md5: 630c3037409fc170eeb6ba05357087aa
+      size: 22164
+  upload_annotation_esim_subgroups@eSA_Scheme:
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/esim_prd_synthetic_qa_with_subgroups --subgroup eSA_Scheme 
+      --working-group eSim --sample-size 1000 --dataset-name-prefix 
+      esim_eSA_Scheme --logger-level DEBUG
+    deps:
+    - path: data/esim_prd_synthetic_qa_with_subgroups
+      hash: md5
+      md5: c67e63493fe27a7cf2d7e59cf28b00bf.dir
+      size: 21361506
+      nfiles: 11
+    - path: gsma_dataset_creation/argilla_cli.py
+      hash: md5
+      md5: cf30c259ec2abc16adad41b33830495b
+      size: 57170
+    - path: gsma_dataset_creation/validation/argilla_subgroup_uploader.py
+      hash: md5
+      md5: 630c3037409fc170eeb6ba05357087aa
+      size: 22164
+  upload_annotation_esim_subgroups@Compliance_Policy:
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/esim_prd_synthetic_qa_with_subgroups --subgroup Compliance_Policy 
+      --working-group eSim --sample-size 1000 --dataset-name-prefix 
+      esim_Compliance_Policy --logger-level DEBUG
+    deps:
+    - path: data/esim_prd_synthetic_qa_with_subgroups
+      hash: md5
+      md5: 1f1f681e66dc2e549cb97eed3525ef36.dir
+      size: 21318946
+      nfiles: 5
+    - path: gsma_dataset_creation/argilla_cli.py
+      hash: md5
+      md5: 96da29c6bf660ee8add034e08fec92ab
+      size: 58419
+    - path: gsma_dataset_creation/validation/argilla_subgroup_uploader.py
+      hash: md5
+      md5: 71199696b1c81567d1da0a9da07391b1
+      size: 22119
+  upload_annotation_esim_subgroups@Certification_Policy:
+    cmd: uv run gsma argilla upload-by-subgroup --dataset-path 
+      data/esim_prd_synthetic_qa_with_subgroups --subgroup Certification_Policy 
+      --working-group eSim --sample-size 1000 --dataset-name-prefix 
+      esim_Certification_Policy --logger-level DEBUG
+    deps:
+    - path: data/esim_prd_synthetic_qa_with_subgroups
+      hash: md5
+      md5: 20ebbdfeeb6d07d58d505189b73b43bf.dir
+      size: 21363226
+      nfiles: 12
+    - path: gsma_dataset_creation/argilla_cli.py
+      hash: md5
+      md5: cf30c259ec2abc16adad41b33830495b
+      size: 57170
+    - path: gsma_dataset_creation/validation/argilla_subgroup_uploader.py
+      hash: md5
+      md5: 630c3037409fc170eeb6ba05357087aa
+      size: 22164
diff --git a/pipelines/annotation/dvc.yaml b/pipelines/annotation/dvc.yaml
index 17cdddf..b780802 100644
--- a/pipelines/annotation/dvc.yaml
+++ b/pipelines/annotation/dvc.yaml
@@ -4,17 +4,17 @@ stages:
     wdir: ../..
     cmd: >-
       uv run gsma add-subgroup-to-dataset
-      --dataset-repo mantisnlp/gsma_prd_synthetic_qa
+      --dataset-repo jamesaidev001/esim_prd_synthetic_qa
       --working-groups data/working_groups_mapping.json
-      --output data/gsma_prd_synthetic_qa_with_subgroups
+      --output data/esim_prd_synthetic_qa_with_subgroups
       --log-level INFO
     deps:
       - data/working_groups_mapping.json
       - gsma_dataset_creation/cli.py
       - gsma_dataset_creation/subgroup_adder.py
     outs:
-      - data/gsma_prd_synthetic_qa_with_subgroups
-    desc: "Add subgroup classification to HuggingFace dataset from Hub (mantisnlp/gsma_prd_synthetic_qa) based on working groups mapping"
+      - data/esim_prd_synthetic_qa_with_subgroups
+    desc: "Add subgroup classification to HuggingFace dataset from Hub (jamesaidev001/esim_prd_synthetic_qa) based on working groups mapping"
 
   # Step 2: Upload working groups without subgroups to Argilla for annotation
   # NOTE: Includes automatic retry with exponential backoff for SQLite lock errors.
@@ -52,26 +52,26 @@ stages:
   # eSIM has multiple subgroups that all share one workspace
   upload_annotation_esim_subgroups:
     foreach:
+      - Certification_Policy
+      - Compliance
+      - EID_Scheme
       - Requirements
+      - Security
       - Technical
       - Testing
-      - Compliance
-      - Security
-      - Certification_policy
-      - Eid_scheme
-      - Esa_scheme
+      - eSA_Scheme
     do:
       wdir: ../..
       cmd: >-
         uv run gsma argilla upload-by-subgroup
-        --dataset-path data/gsma_prd_synthetic_qa_with_subgroups
+        --dataset-path data/esim_prd_synthetic_qa_with_subgroups
         --subgroup ${item}
-        --working-group eSIM
+        --working-group eSim
         --sample-size 1000
-        --dataset-name-prefix gsma_annotation_esim_${item}
-        --logger-level INFO
+        --dataset-name-prefix esim_${item}
+        --logger-level DEBUG
       deps:
-        - data/gsma_prd_synthetic_qa_with_subgroups
+        - data/esim_prd_synthetic_qa_with_subgroups
         - gsma_dataset_creation/argilla_cli.py
         - gsma_dataset_creation/validation/argilla_subgroup_uploader.py
       desc: "Upload 1000 Q&A pairs from eSIM/${item} subgroup to Argilla for quality annotation (eSIM workspace)"
diff --git a/pyproject.toml b/pyproject.toml
index 53d62c5..cc52fd7 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [project]
 name = "gsma-dataset-creation"
-version = "0.1.0"
+version = "0.1.1"
 description = "GSMA document processing pipeline"
 readme = "README.md"
 requires-python = ">=3.11"