diff --git a/metagpt/ext/sela/README.md b/metagpt/ext/sela/README.md
index 3fa03ee08..829306e36 100644
--- a/metagpt/ext/sela/README.md
+++ b/metagpt/ext/sela/README.md
@@ -213,7 +213,7 @@ #### Run
 The `log` folder will contain the experimental configuration and the generated scheme, and the `workspace` folder will save the final results generated by aide
 
 ```
-python experimenter/aide.py
+python runner/aide.py
 ```
 
 ### Autogluon
diff --git a/metagpt/ext/sela/data/custom_task.py b/metagpt/ext/sela/data/custom_task.py
index 3371d5b1c..08a7cbabb 100644
--- a/metagpt/ext/sela/data/custom_task.py
+++ b/metagpt/ext/sela/data/custom_task.py
@@ -1,7 +1,7 @@
 import os
 
 from metagpt.ext.sela.data.dataset import SPECIAL_INSTRUCTIONS
-from metagpt.ext.sela.experimenter.mle_bench.instructions import (
+from metagpt.ext.sela.runner.mle_bench.instructions import (
     ADDITIONAL_NOTES,
     INSTRUCTIONS,
     INSTRUCTIONS_OBFUSCATED,
diff --git a/metagpt/ext/sela/research_assistant.py b/metagpt/ext/sela/experimenter.py
similarity index 100%
rename from metagpt/ext/sela/research_assistant.py
rename to metagpt/ext/sela/experimenter.py
diff --git a/metagpt/ext/sela/run_experiment.py b/metagpt/ext/sela/run_experiment.py
index 4cced19c3..32130a6fb 100644
--- a/metagpt/ext/sela/run_experiment.py
+++ b/metagpt/ext/sela/run_experiment.py
@@ -2,12 +2,12 @@ import argparse
 import asyncio
 
 from metagpt.ext.sela.data.custom_task import get_mle_is_lower_better, get_mle_task_id
-from metagpt.ext.sela.experimenter.autogluon import GluonExperimenter
-from metagpt.ext.sela.experimenter.autosklearn import AutoSklearnExperimenter
-from metagpt.ext.sela.experimenter.custom import CustomExperimenter
-from metagpt.ext.sela.experimenter.experimenter import Experimenter
-from metagpt.ext.sela.experimenter.mcts import MCTSExperimenter
-from metagpt.ext.sela.experimenter.random_search import RandomSearchExperimenter
+from metagpt.ext.sela.runner.autogluon import GluonRunner
+from metagpt.ext.sela.runner.autosklearn import AutoSklearnRunner
+from metagpt.ext.sela.runner.custom import CustomRunner
+from metagpt.ext.sela.runner.mcts import MCTSRunner
+from metagpt.ext.sela.runner.random_search import RandomSearchRunner
+from metagpt.ext.sela.runner.runner import Runner
 
 
 def get_args(cmd=True):
@@ -74,24 +74,24 @@ def get_di_args(parser):
 
 async def main(args):
     if args.exp_mode == "mcts":
-        experimenter = MCTSExperimenter(args)
+        runner = MCTSRunner(args)
     elif args.exp_mode == "greedy":
-        experimenter = MCTSExperimenter(args, tree_mode="greedy")
+        runner = MCTSRunner(args, tree_mode="greedy")
     elif args.exp_mode == "random":
-        experimenter = MCTSExperimenter(args, tree_mode="random")
+        runner = MCTSRunner(args, tree_mode="random")
     elif args.exp_mode == "rs":
-        experimenter = RandomSearchExperimenter(args)
+        runner = RandomSearchRunner(args)
     elif args.exp_mode == "base":
-        experimenter = Experimenter(args)
+        runner = Runner(args)
     elif args.exp_mode == "autogluon":
-        experimenter = GluonExperimenter(args)
+        runner = GluonRunner(args)
     elif args.exp_mode == "custom":
-        experimenter = CustomExperimenter(args)
+        runner = CustomRunner(args)
     elif args.exp_mode == "autosklearn":
-        experimenter = AutoSklearnExperimenter(args)
+        runner = AutoSklearnRunner(args)
     else:
         raise ValueError(f"Invalid exp_mode: {args.exp_mode}")
-    await experimenter.run_experiment()
+    await runner.run_experiment()
 
 
 if __name__ == "__main__":
diff --git a/metagpt/ext/sela/experimenter/__init__.py b/metagpt/ext/sela/runner/__init__.py
similarity index 100%
rename from metagpt/ext/sela/experimenter/__init__.py
rename to metagpt/ext/sela/runner/__init__.py
diff --git a/metagpt/ext/sela/experimenter/aide.py b/metagpt/ext/sela/runner/aide.py
similarity index 100%
rename from metagpt/ext/sela/experimenter/aide.py
rename to metagpt/ext/sela/runner/aide.py
diff --git a/metagpt/ext/sela/experimenter/autogluon.py b/metagpt/ext/sela/runner/autogluon.py
similarity index 98%
rename from metagpt/ext/sela/experimenter/autogluon.py
rename to metagpt/ext/sela/runner/autogluon.py
index f547ce4ba..48737da04 100644
--- a/metagpt/ext/sela/experimenter/autogluon.py
+++ b/metagpt/ext/sela/runner/autogluon.py
@@ -3,7 +3,7 @@ from datetime import datetime
 
 import pandas as pd
 
-from metagpt.ext.sela.experimenter.custom import CustomExperimenter
+from metagpt.ext.sela.runner.custom import CustomRunner
 
 
 class AGRunner:
@@ -102,7 +102,7 @@ class AGRunner:
         return train_data, dev_data, dev_wo_target_data, test_data
 
 
-class GluonExperimenter(CustomExperimenter):
+class GluonRunner(CustomRunner):
     result_path: str = "results/autogluon"
 
     def __init__(self, args, **kwargs):
diff --git a/metagpt/ext/sela/experimenter/autosklearn.py b/metagpt/ext/sela/runner/autosklearn.py
similarity index 96%
rename from metagpt/ext/sela/experimenter/autosklearn.py
rename to metagpt/ext/sela/runner/autosklearn.py
index f6ff267e7..7d0eb364e 100644
--- a/metagpt/ext/sela/experimenter/autosklearn.py
+++ b/metagpt/ext/sela/runner/autosklearn.py
@@ -4,7 +4,7 @@ from functools import partial
 import pandas as pd
 
 from metagpt.ext.sela.evaluation.evaluation import evaluate_score
-from metagpt.ext.sela.experimenter.custom import CustomExperimenter
+from metagpt.ext.sela.runner.custom import CustomRunner
 
 
 def custom_scorer(y_true, y_pred, metric_name):
@@ -69,7 +69,7 @@ class ASRunner:
         return {"test_preds": test_preds, "dev_preds": dev_preds}
 
 
-class AutoSklearnExperimenter(CustomExperimenter):
+class AutoSklearnRunner(CustomRunner):
     result_path: str = "results/autosklearn"
 
     def __init__(self, args, **kwargs):
diff --git a/metagpt/ext/sela/experimenter/custom.py b/metagpt/ext/sela/runner/custom.py
similarity index 95%
rename from metagpt/ext/sela/experimenter/custom.py
rename to metagpt/ext/sela/runner/custom.py
index 70df1a78e..e9a8ee276 100644
--- a/metagpt/ext/sela/experimenter/custom.py
+++ b/metagpt/ext/sela/runner/custom.py
@@ -3,11 +3,11 @@ import os
 import pandas as pd
 
 from metagpt.ext.sela.evaluation.evaluation import evaluate_score
-from metagpt.ext.sela.experimenter.experimenter import Experimenter
+from metagpt.ext.sela.runner.runner import Runner
 from metagpt.ext.sela.search.tree_search import create_initial_state
 
 
-class CustomExperimenter(Experimenter):
+class CustomRunner(Runner):
     result_path: str = "results/custom"
 
     def __init__(self, args, **kwargs):
diff --git a/metagpt/ext/sela/experimenter/mcts.py b/metagpt/ext/sela/runner/mcts.py
similarity index 96%
rename from metagpt/ext/sela/experimenter/mcts.py
rename to metagpt/ext/sela/runner/mcts.py
index 9fd66121d..8b6c14100 100644
--- a/metagpt/ext/sela/experimenter/mcts.py
+++ b/metagpt/ext/sela/runner/mcts.py
@@ -5,11 +5,11 @@ from metagpt.ext.sela.evaluation.evaluation import (
     node_evaluate_score_sela,
 )
 from metagpt.ext.sela.evaluation.visualize_mcts import get_tree_text
-from metagpt.ext.sela.experimenter.experimenter import Experimenter
+from metagpt.ext.sela.runner.runner import Runner
 from metagpt.ext.sela.search.search_algorithm import MCTS, Greedy, Random
 
 
-class MCTSExperimenter(Experimenter):
+class MCTSRunner(Runner):
     result_path: str = "results/mcts"
 
     def __init__(self, args, tree_mode=None, **kwargs):
diff --git a/metagpt/ext/sela/experimenter/mle_bench/instructions.py b/metagpt/ext/sela/runner/mle_bench/instructions.py
similarity index 98%
rename from metagpt/ext/sela/experimenter/mle_bench/instructions.py
rename to metagpt/ext/sela/runner/mle_bench/instructions.py
index 5c9c7b70f..136726bcf 100644
--- a/metagpt/ext/sela/experimenter/mle_bench/instructions.py
+++ b/metagpt/ext/sela/runner/mle_bench/instructions.py
@@ -8,7 +8,7 @@ You will be given a machine learning task. You must solve the task by training a
 
 BENCHMARK INSTRUCTIONS
 ------
-First, please read the instructions below which are specific to our benchmark setup. Note, these instructions can be found again in `experimenter/mle_bench/instructions.py`.
+First, please read the instructions below which are specific to our benchmark setup. Note, these instructions can be found again in `runner/mle_bench/instructions.py`.
 
 - **Competition-specific instructions**: The instructions for the competition are available in `{dataset_dir}/description.md`. Read this file to understand the problem setup for the competition you are currently solving.
 - **Dataset**: The dataset files for the competition are available in `{dataset_dir}`.
@@ -28,7 +28,7 @@ You will be given a machine learning task. You must solve the task by training a
 
 BENCHMARK INSTRUCTIONS
 ------
-First, please read the instructions below which are specific to our benchmark setup. Note, these instructions can be found again in `experimenter/mle_bench/instructions.py`.
+First, please read the instructions below which are specific to our benchmark setup. Note, these instructions can be found again in `runner/mle_bench/instructions.py`.
 
 - **Task-specific instructions**: The instructions for the task are available in `{dataset_dir}/description.md`. Read this file to understand the problem setup for the task you are currently solving.
 - **Dataset**: The dataset files for the task are available in `{dataset_dir}/`.
diff --git a/metagpt/ext/sela/experimenter/random_search.py b/metagpt/ext/sela/runner/random_search.py
similarity index 92%
rename from metagpt/ext/sela/experimenter/random_search.py
rename to metagpt/ext/sela/runner/random_search.py
index 5617ee601..8ce42f0ff 100644
--- a/metagpt/ext/sela/experimenter/random_search.py
+++ b/metagpt/ext/sela/runner/random_search.py
@@ -1,6 +1,6 @@
-from metagpt.ext.sela.experimenter.experimenter import Experimenter
+from metagpt.ext.sela.experimenter import ResearchAssistant
 from metagpt.ext.sela.insights.instruction_generator import InstructionGenerator
-from metagpt.ext.sela.research_assistant import ResearchAssistant
+from metagpt.ext.sela.runner.runner import Runner
 from metagpt.ext.sela.utils import get_exp_pool_path
 
 EXPS_PROMPT = """
@@ -10,7 +10,7 @@ When doing the tasks, you can refer to the insights below:
 """
 
 
-class RandomSearchExperimenter(Experimenter):
+class RandomSearchRunner(Runner):
     result_path: str = "results/random_search"
 
     async def run_experiment(self):
diff --git a/metagpt/ext/sela/experimenter/experimenter.py b/metagpt/ext/sela/runner/runner.py
similarity index 98%
rename from metagpt/ext/sela/experimenter/experimenter.py
rename to metagpt/ext/sela/runner/runner.py
index 3df46b74b..7ab83c6c3 100644
--- a/metagpt/ext/sela/experimenter/experimenter.py
+++ b/metagpt/ext/sela/runner/runner.py
@@ -6,12 +6,12 @@ import numpy as np
 import pandas as pd
 
 from metagpt.ext.sela.evaluation.evaluation import evaluate_score
-from metagpt.ext.sela.research_assistant import ResearchAssistant
+from metagpt.ext.sela.experimenter import ResearchAssistant
 from metagpt.ext.sela.search.tree_search import create_initial_state
 from metagpt.ext.sela.utils import DATA_CONFIG, save_notebook
 
 
-class Experimenter:
+class Runner:
     result_path: str = "results/base"
     data_config = DATA_CONFIG
     start_task_id = 1
diff --git a/metagpt/ext/sela/search/tree_search.py b/metagpt/ext/sela/search/tree_search.py
index cde8dc82a..684426fe6 100644
--- a/metagpt/ext/sela/search/tree_search.py
+++ b/metagpt/ext/sela/search/tree_search.py
@@ -15,8 +15,8 @@ from metagpt.ext.sela.data.dataset import (
     get_split_dataset_path,
 )
 from metagpt.ext.sela.evaluation.evaluation import evaluate_score
+from metagpt.ext.sela.experimenter import ResearchAssistant, TimeoutException
 from metagpt.ext.sela.insights.instruction_generator import InstructionGenerator
-from metagpt.ext.sela.research_assistant import ResearchAssistant, TimeoutException
 from metagpt.ext.sela.utils import get_exp_pool_path, load_execute_notebook, mcts_logger
 from metagpt.tools.tool_recommend import ToolRecommender
 from metagpt.utils.common import read_json_file