diff --git a/examples/mnist-keras/client/get_data.py b/examples/mnist-keras/client/data.py
old mode 100755
new mode 100644
similarity index 57%
rename from examples/mnist-keras/client/get_data.py
rename to examples/mnist-keras/client/data.py
index ed123a4a3..749722d98
--- a/examples/mnist-keras/client/get_data.py
+++ b/examples/mnist-keras/client/data.py
@@ -4,6 +4,52 @@
 import numpy as np
 import tensorflow as tf
 
+dir_path = os.path.dirname(os.path.realpath(__file__))
+abs_path = os.path.abspath(dir_path)
+
+NUM_CLASSES = 10
+
+
+def get_data(out_dir="data"):
+    # Make dir if necessary
+    if not os.path.exists(out_dir):
+        os.mkdir(out_dir)
+
+    # Download data
+    (x_train, y_train), (x_test, y_test) = tf.keras.datasets.mnist.load_data()
+    np.savez(f"{out_dir}/mnist.npz", x_train=x_train, y_train=y_train, x_test=x_test, y_test=y_test)
+
+
+def load_data(data_path, is_train=True):
+    """Load data from disk.
+
+    :param data_path: Path to data file.
+    :type data_path: str
+    :param is_train: Whether to load train or test data.
+    :type is_train: bool
+    :return: Tuple of data and labels.
+    :rtype: tuple
+    """
+    if data_path is None:
+        data_path = os.environ.get("FEDN_DATA_PATH", abs_path + "/data/clients/1/mnist.npz")
+
+    data = np.load(data_path)
+
+    if is_train:
+        X = data["x_train"]
+        y = data["y_train"]
+    else:
+        X = data["x_test"]
+        y = data["y_test"]
+
+    # Normalize
+    X = X.astype("float32")
+    X = np.expand_dims(X, -1)
+    X = X / 255
+    y = tf.keras.utils.to_categorical(y, NUM_CLASSES)
+
+    return X, y
+
 
 def splitset(dataset, parts):
     n = dataset.shape[0]
@@ -33,16 +79,8 @@ def split(dataset="data/mnist.npz", outdir="data", n_splits=2):
         np.savez(f"{subdir}/mnist.npz", x_train=data["x_train"][i], y_train=data["y_train"][i], x_test=data["x_test"][i], y_test=data["y_test"][i])
 
 
-def get_data(out_dir="data"):
-    # Make dir if necessary
-    if not os.path.exists(out_dir):
-        os.mkdir(out_dir)
-
-    # Download data
-    (x_train, y_train), (x_test, y_test) = tf.keras.datasets.mnist.load_data()
-    np.savez(f"{out_dir}/mnist.npz", x_train=x_train, y_train=y_train, x_test=x_test, y_test=y_test)
-
-
 if __name__ == "__main__":
-    get_data()
-    split()
+    # Prepare data if not already done
+    if not os.path.exists(abs_path + "/data/clients/1"):
+        get_data()
+        split()
diff --git a/examples/mnist-keras/client/entrypoint.py b/examples/mnist-keras/client/entrypoint.py
deleted file mode 100755
index 1ed8f2f77..000000000
--- a/examples/mnist-keras/client/entrypoint.py
+++ /dev/null
@@ -1,194 +0,0 @@
-import json
-import os
-
-import fire
-import numpy as np
-import tensorflow as tf
-
-from fedn.utils.helpers.helpers import get_helper, save_metadata, save_metrics
-
-HELPER_MODULE = "numpyhelper"
-helper = get_helper(HELPER_MODULE)
-
-NUM_CLASSES = 10
-
-dir_path = os.path.dirname(os.path.realpath(__file__))
-abs_path = os.path.abspath(dir_path)
-
-
-def _get_data_path():
-    data_path = os.environ.get("FEDN_DATA_PATH", abs_path + "/data/clients/1/mnist.npz")
-
-    return data_path
-
-
-def compile_model(img_rows=28, img_cols=28):
-    """Compile the TF model.
-
-    param: img_rows: The number of rows in the image
-    type: img_rows: int
-    param: img_cols: The number of rows in the image
-    type: img_cols: int
-    return: The compiled model
-    type: keras.model.Sequential
-    """
-    # Set input shape
-    input_shape = (img_rows, img_cols, 1)
-
-    # Define model
-    model = tf.keras.models.Sequential()
-    model.add(tf.keras.layers.Flatten(input_shape=input_shape))
-    model.add(tf.keras.layers.Dense(64, activation="relu"))
-    model.add(tf.keras.layers.Dropout(0.5))
-    model.add(tf.keras.layers.Dense(32, activation="relu"))
-    model.add(tf.keras.layers.Dense(NUM_CLASSES, activation="softmax"))
-    model.compile(loss=tf.keras.losses.categorical_crossentropy, optimizer=tf.keras.optimizers.Adam(), metrics=["accuracy"])
-
-    return model
-
-
-def load_data(data_path, is_train=True):
-    # Load data
-    if data_path is None:
-        data = np.load(_get_data_path())
-    else:
-        data = np.load(data_path)
-
-    if is_train:
-        X = data["x_train"]
-        y = data["y_train"]
-    else:
-        X = data["x_test"]
-        y = data["y_test"]
-
-    # Normalize
-    X = X.astype("float32")
-    X = np.expand_dims(X, -1)
-    X = X / 255
-    y = tf.keras.utils.to_categorical(y, NUM_CLASSES)
-
-    return X, y
-
-
-def init_seed(out_path="../seed.npz"):
-    """Initialize seed model and save it to file.
-
-    :param out_path: The path to save the seed model to.
-    :type out_path: str
-    """
-    weights = compile_model().get_weights()
-    helper.save(weights, out_path)
-
-
-def train(in_model_path, out_model_path, data_path=None, batch_size=32, epochs=1):
-    """Complete a model update.
-
-    Load model paramters from in_model_path (managed by the FEDn client),
-    perform a model update, and write updated paramters
-    to out_model_path (picked up by the FEDn client).
-
-    :param in_model_path: The path to the input model.
-    :type in_model_path: str
-    :param out_model_path: The path to save the output model to.
-    :type out_model_path: str
-    :param data_path: The path to the data file.
-    :type data_path: str
-    :param batch_size: The batch size to use.
-    :type batch_size: int
-    :param epochs: The number of epochs to train.
-    :type epochs: int
-    """
-    # Load data
-    x_train, y_train = load_data(data_path)
-
-    # Load model
-    model = compile_model()
-    weights = helper.load(in_model_path)
-    model.set_weights(weights)
-
-    # Train
-    model.fit(x_train, y_train, batch_size=batch_size, epochs=epochs)
-
-    # Metadata needed for aggregation server side
-    metadata = {
-        # num_examples are mandatory
-        "num_examples": len(x_train),
-        "batch_size": batch_size,
-        "epochs": epochs,
-    }
-
-    # Save JSON metadata file (mandatory)
-    save_metadata(metadata, out_model_path)
-
-    # Save model update (mandatory)
-    weights = model.get_weights()
-    helper.save(weights, out_model_path)
-
-
-def validate(in_model_path, out_json_path, data_path=None):
-    """Validate model.
-
-    :param in_model_path: The path to the input model.
-    :type in_model_path: str
-    :param out_json_path: The path to save the output JSON to.
-    :type out_json_path: str
-    :param data_path: The path to the data file.
-    :type data_path: str
-    """
-    # Load data
-    x_train, y_train = load_data(data_path)
-    x_test, y_test = load_data(data_path, is_train=False)
-
-    # Load model
-    model = compile_model()
-    helper = get_helper(HELPER_MODULE)
-    weights = helper.load(in_model_path)
-    model.set_weights(weights)
-
-    # Evaluate
-    model_score = model.evaluate(x_train, y_train)
-    model_score_test = model.evaluate(x_test, y_test)
-    y_pred = model.predict(x_test)
-    y_pred = np.argmax(y_pred, axis=1)
-
-    # JSON schema
-    report = {
-        "training_loss": model_score[0],
-        "training_accuracy": model_score[1],
-        "test_loss": model_score_test[0],
-        "test_accuracy": model_score_test[1],
-    }
-
-    # Save JSON
-    save_metrics(report, out_json_path)
-
-
-def predict(in_model_path, out_json_path, data_path=None):
-    # Using test data for inference but another dataset could be loaded
-    x_test, _ = load_data(data_path, is_train=False)
-
-    # Load model
-    model = compile_model()
-    helper = get_helper(HELPER_MODULE)
-    weights = helper.load(in_model_path)
-    model.set_weights(weights)
-
-    # Infer
-    y_pred = model.predict(x_test)
-    y_pred = np.argmax(y_pred, axis=1)
-
-    # Save JSON
-    with open(out_json_path, "w") as fh:
-        fh.write(json.dumps({"predictions": y_pred.tolist()}))
-
-
-if __name__ == "__main__":
-    fire.Fire(
-        {
-            "init_seed": init_seed,
-            "train": train,
-            "validate": validate,
-            "predict": predict,
-            "_get_data_path": _get_data_path,  # for testing
-        }
-    )
diff --git a/examples/mnist-keras/client/fedn.yaml b/examples/mnist-keras/client/fedn.yaml
index 776dc9594..30873488b 100644
--- a/examples/mnist-keras/client/fedn.yaml
+++ b/examples/mnist-keras/client/fedn.yaml
@@ -1,12 +1,12 @@
 python_env: python_env.yaml
 entry_points:
   build:
-    command: python entrypoint.py init_seed
+    command: python model.py
   startup:
-    command: python get_data.py
+    command: python data.py
   train:
-    command: python entrypoint.py train $ENTRYPOINT_OPTS
+    command: python train.py
   validate:
-    command: python entrypoint.py validate $ENTRYPOINT_OPTS
+    command: python validate.py
   predict:
-    command: python entrypoint.py predict $ENTRYPOINT_OPTS
+    command: python predict.py
\ No newline at end of file
diff --git a/examples/mnist-keras/client/model.py b/examples/mnist-keras/client/model.py
new file mode 100644
index 000000000..696ac8c13
--- /dev/null
+++ b/examples/mnist-keras/client/model.py
@@ -0,0 +1,71 @@
+import tensorflow as tf
+
+from fedn.utils.helpers.helpers import get_helper
+
+NUM_CLASSES = 10
+HELPER_MODULE = "numpyhelper"
+helper = get_helper(HELPER_MODULE)
+
+
+def compile_model(img_rows=28, img_cols=28):
+    """Compile the TF model.
+
+    param: img_rows: The number of rows in the image
+    type: img_rows: int
+    param: img_cols: The number of rows in the image
+    type: img_cols: int
+    return: The compiled model
+    type: keras.model.Sequential
+    """
+    # Set input shape
+    input_shape = (img_rows, img_cols, 1)
+
+    # Define model
+    model = tf.keras.models.Sequential()
+    model.add(tf.keras.layers.Flatten(input_shape=input_shape))
+    model.add(tf.keras.layers.Dense(64, activation="relu"))
+    model.add(tf.keras.layers.Dropout(0.5))
+    model.add(tf.keras.layers.Dense(32, activation="relu"))
+    model.add(tf.keras.layers.Dense(NUM_CLASSES, activation="softmax"))
+    model.compile(loss=tf.keras.losses.categorical_crossentropy, optimizer=tf.keras.optimizers.Adam(), metrics=["accuracy"])
+    return model
+
+
+def save_parameters(model, out_path):
+    """Save model parameters to file.
+
+    :param model: The model to serialize.
+    :type model: keras.model.Sequential
+    :param out_path: The path to save the model to.
+    :type out_path: str
+    """
+    weights = model.get_weights()
+    helper.save(weights, out_path)
+
+
+def load_parameters(model_path):
+    """Load model parameters from file and populate model.
+
+    :param model_path: The path to load from.
+    :type model_path: str
+    :return: The loaded model.
+    :rtype: keras.model.Sequential
+    """
+    model = compile_model()
+    weights = helper.load(model_path)
+    model.set_weights(weights)
+    return model
+
+
+def init_seed(out_path="../seed.npz"):
+    """Initialize seed model and save it to file.
+
+    :param out_path: The path to save the seed model to.
+    :type out_path: str
+    """
+    weights = compile_model().get_weights()
+    helper.save(weights, out_path)
+
+
+if __name__ == "__main__":
+    init_seed("../seed.npz")
diff --git a/examples/mnist-keras/client/predict.py b/examples/mnist-keras/client/predict.py
new file mode 100644
index 000000000..9d502ed75
--- /dev/null
+++ b/examples/mnist-keras/client/predict.py
@@ -0,0 +1,30 @@
+import json
+import os
+import sys
+
+import numpy as np
+from data import load_data
+from model import load_parameters
+
+dir_path = os.path.dirname(os.path.realpath(__file__))
+sys.path.append(os.path.abspath(dir_path))
+
+
+def predict(in_model_path, out_json_path, data_path=None):
+    # Using test data for inference but another dataset could be loaded
+    x_test, _ = load_data(data_path, is_train=False)
+
+    # Load model
+    model = load_parameters(in_model_path)
+
+    # Infer
+    y_pred = model.predict(x_test)
+    y_pred = np.argmax(y_pred, axis=1)
+
+    # Save JSON
+    with open(out_json_path, "w") as fh:
+        fh.write(json.dumps({"predictions": y_pred.tolist()}))
+
+
+if __name__ == "__main__":
+    predict(sys.argv[1], sys.argv[2])
diff --git a/examples/mnist-keras/client/train.py b/examples/mnist-keras/client/train.py
new file mode 100644
index 000000000..e899f7c4d
--- /dev/null
+++ b/examples/mnist-keras/client/train.py
@@ -0,0 +1,56 @@
+import os
+import sys
+
+from data import load_data
+from model import load_parameters, save_parameters
+
+from fedn.utils.helpers.helpers import save_metadata
+
+dir_path = os.path.dirname(os.path.realpath(__file__))
+sys.path.append(os.path.abspath(dir_path))
+
+
+def train(in_model_path, out_model_path, data_path=None, batch_size=32, epochs=1):
+    """Complete a model update.
+
+    Load model paramters from in_model_path (managed by the FEDn client),
+    perform a model update, and write updated paramters
+    to out_model_path (picked up by the FEDn client).
+
+    :param in_model_path: The path to the input model.
+    :type in_model_path: str
+    :param out_model_path: The path to save the output model to.
+    :type out_model_path: str
+    :param data_path: The path to the data file.
+    :type data_path: str
+    :param batch_size: The batch size to use.
+    :type batch_size: int
+    :param epochs: The number of epochs to train.
+    :type epochs: int
+    """
+    # Load data
+    x_train, y_train = load_data(data_path)
+
+    # Load model
+    model = load_parameters(in_model_path)
+
+    # Train
+    model.fit(x_train, y_train, batch_size=batch_size, epochs=epochs)
+
+    # Metadata needed for aggregation server side
+    metadata = {
+        # num_examples are mandatory
+        "num_examples": len(x_train),
+        "batch_size": batch_size,
+        "epochs": epochs,
+    }
+
+    # Save JSON metadata file (mandatory)
+    save_metadata(metadata, out_model_path)
+
+    # Save model update (mandatory)
+    save_parameters(model, out_model_path)
+
+
+if __name__ == "__main__":
+    train(sys.argv[1], sys.argv[2])
diff --git a/examples/mnist-keras/client/validate.py b/examples/mnist-keras/client/validate.py
new file mode 100644
index 000000000..6b462dfc2
--- /dev/null
+++ b/examples/mnist-keras/client/validate.py
@@ -0,0 +1,50 @@
+import os
+import sys
+
+import numpy as np
+from data import load_data
+from model import load_parameters
+
+from fedn.utils.helpers.helpers import save_metrics
+
+dir_path = os.path.dirname(os.path.realpath(__file__))
+sys.path.append(os.path.abspath(dir_path))
+
+
+def validate(in_model_path, out_json_path, data_path=None):
+    """Validate model.
+
+    :param in_model_path: The path to the input model.
+    :type in_model_path: str
+    :param out_json_path: The path to save the output JSON to.
+    :type out_json_path: str
+    :param data_path: The path to the data file.
+    :type data_path: str
+    """
+    # Load data
+    x_train, y_train = load_data(data_path)
+    x_test, y_test = load_data(data_path, is_train=False)
+
+    # Load model
+    model = load_parameters(in_model_path)
+
+    # Evaluate
+    model_score = model.evaluate(x_train, y_train)
+    model_score_test = model.evaluate(x_test, y_test)
+    y_pred = model.predict(x_test)
+    y_pred = np.argmax(y_pred, axis=1)
+
+    # JSON schema
+    report = {
+        "training_loss": model_score[0],
+        "training_accuracy": model_score[1],
+        "test_loss": model_score_test[0],
+        "test_accuracy": model_score_test[1],
+    }
+
+    # Save JSON
+    save_metrics(report, out_json_path)
+
+
+if __name__ == "__main__":
+    validate(sys.argv[1], sys.argv[2])