diff --git a/software/config.yaml b/software/config.yaml
index 7e6c3c6..c647beb 100644
--- a/software/config.yaml
+++ b/software/config.yaml
@@ -1,7 +1,7 @@
 client:
   enabled: false
   url: null
-  platform: auto
+  platform: ''
 
 llm:
   service: litellm
@@ -13,7 +13,7 @@ llm:
   temperature: 0.8
 
 local:
-  enabled: true
+  enabled: false
   tts_service: piper
   stt_service: local-whisper
 
diff --git a/software/pyproject.toml b/software/pyproject.toml
index e85bad0..8ff54ac 100644
--- a/software/pyproject.toml
+++ b/software/pyproject.toml
@@ -51,6 +51,7 @@ pytest = "^8.1.1"
 target-version = ['py311']
 
 [tool.isort]
-profile = "black"
-multi_line_output = 3
 include_trailing_comma = true
+known_first_party = ["source"]
+multi_line_output = 3
+profile = "black"
diff --git a/software/source/__init__.py b/software/source/__init__.py
index b6bac92..64ec270 100644
--- a/software/source/__init__.py
+++ b/software/source/__init__.py
@@ -1,6 +1,11 @@
+"""
+Loads environment variables and creates a global configuration object.
+"""
+
 from dotenv import load_dotenv
-from source.core.models import Config
+
+from source.core.config import Config, get_config
 
 load_dotenv()
 
-config = Config()
+config: Config = get_config()
diff --git a/software/source/core/config.py b/software/source/core/config.py
new file mode 100644
index 0000000..1444031
--- /dev/null
+++ b/software/source/core/config.py
@@ -0,0 +1,110 @@
+"""
+Application configuration model.
+"""
+
+import os
+from functools import lru_cache
+from typing import Any
+
+from pydantic_settings import (
+    BaseSettings,
+    DotEnvSettingsSource,
+    PydanticBaseSettingsSource,
+    SettingsConfigDict,
+    YamlConfigSettingsSource,
+)
+
+from source.core.models import LLM, STT, TTS, Client, Local, Server, Tunnel
+
+APP_PREFIX: str = os.getenv("01_PREFIX", "01_")
+
+
+class Config(BaseSettings):
+    """
+    Base configuration model
+    """
+
+    client: Client = Client()
+    llm: LLM = LLM()
+    local: Local = Local()
+    server: Server = Server()
+    stt: STT = STT()
+    tts: TTS = TTS()
+    tunnel: Tunnel = Tunnel()
+
+    model_config = SettingsConfigDict(extra="allow")
+
+    @classmethod
+    def settings_customise_sources(
+        cls,
+        settings_cls: type[BaseSettings],
+        init_settings: PydanticBaseSettingsSource,
+        env_settings: PydanticBaseSettingsSource,
+        dotenv_settings: PydanticBaseSettingsSource,
+        file_secret_settings: PydanticBaseSettingsSource,
+    ) -> tuple[PydanticBaseSettingsSource, ...]:
+        """
+        Modify the order of precedence for settings sources.
+        """
+        return (
+            DotEnvSettingsSource(
+                settings_cls,
+                env_prefix=APP_PREFIX,
+                env_file=".env",
+                env_file_encoding="utf-8",
+                env_nested_delimiter="_",
+            ),
+            YamlConfigSettingsSource(
+                settings_cls,
+                yaml_file=os.getenv(f"{APP_PREFIX}CONFIG_FILE", "config.yaml"),
+            ),
+        )
+
+    def apply_cli_args(self, args: dict) -> None:
+        """
+        Apply CLI arguments to config.
+        """
+        mapping: dict[str, str] = {
+            "server": "server.enabled",
+            "server_host": "server.host",
+            "server_port": "server.port",
+            "tunnel_service": "tunnel.service",
+            "expose": "tunnel.exposed",
+            "client": "client.enabled",
+            "server_url": "client.url",
+            "client_type": "client.platform",
+            "llm_service": "llm.service",
+            "model": "llm.model",
+            "llm_supports_vision": "llm.vision_enabled",
+            "llm_supports_functions": "llm.functions_enabled",
+            "context_window": "llm.context_window",
+            "max_tokens": "llm.max_tokens",
+            "temperature": "llm.temperature",
+            "tts_service": "tts.service",
+            "stt_service": "stt.service",
+            "local": "local.enabled",
+        }
+
+        for key, path in mapping.items():
+            if key in args and args[key] is not None:
+                self.set_field(path, args[key])
+
+    def set_field(self, field: str, value: Any) -> None:
+        """
+        Set field value
+        """
+        obj: Any = self
+        parts: list[str] = field.split(".")
+
+        for part in parts[:-1]:
+            obj: Any = getattr(obj, part)
+
+        setattr(obj, parts[-1], value)
+
+
+@lru_cache()
+def get_config() -> Config:
+    """
+    Return the application configuration.
+    """
+    return Config()
diff --git a/software/source/core/models.py b/software/source/core/models.py
index 10b871f..e715557 100644
--- a/software/source/core/models.py
+++ b/software/source/core/models.py
@@ -1,17 +1,8 @@
 """
-Application configuration models.
+Application models.
 """
 
-import os
-
 from pydantic import BaseModel
-from pydantic_settings import (
-    BaseSettings,
-    PydanticBaseSettingsSource,
-    YamlConfigSettingsSource,
-)
-
-APP_PREFIX: str = os.getenv("01_APP_PREFIX", "01_")
 
 
 class Client(BaseModel):
@@ -20,8 +11,8 @@ class Client(BaseModel):
     """
 
     enabled: bool = False
-    url: None | str = None
-    platform: str = "auto"
+    url: str | None = None
+    platform: str | None = None
 
 
 class LLM(BaseModel):
@@ -44,8 +35,6 @@ class Local(BaseModel):
     """
 
     enabled: bool = False
-    tts_service: str = "piper"
-    stt_service: str = "local-whisper"
 
 
 class Server(BaseModel):
@@ -81,36 +70,3 @@ class Tunnel(BaseModel):
 
     service: str = "ngrok"
     exposed: bool = False
-
-
-class Config(BaseSettings):
-    """
-    Base configuration model
-    """
-
-    client: Client = Client()
-    llm: LLM = LLM()
-    local: Local = Local()
-    server: Server = Server()
-    stt: STT = STT()
-    tts: TTS = TTS()
-    tunnel: Tunnel = Tunnel()
-
-    @classmethod
-    def settings_customise_sources(
-        cls,
-        settings_cls: type[BaseSettings],
-        init_settings: PydanticBaseSettingsSource,
-        env_settings: PydanticBaseSettingsSource,
-        dotenv_settings: PydanticBaseSettingsSource,
-        file_secret_settings: PydanticBaseSettingsSource,
-    ) -> tuple[PydanticBaseSettingsSource, ...]:
-        """
-        Modify the order of precedence for settings sources.
-        """
-        return (
-            YamlConfigSettingsSource(
-                settings_cls,
-                yaml_file=os.getenv(f"{APP_PREFIX}CONFIG_FILE", "config.yaml"),
-            ),
-        )
diff --git a/software/source/core/utils.py b/software/source/core/utils.py
deleted file mode 100644
index bc09916..0000000
--- a/software/source/core/utils.py
+++ /dev/null
@@ -1,99 +0,0 @@
-"""
-Core utilty functions for the server and client
-"""
-
-import asyncio
-import importlib
-import os
-import platform
-from threading import Thread
-from typing import NoReturn
-
-from source.server.server import main
-from source.server.tunnel import create_tunnel
-
-
-def get_client_platform(config) -> None:
-    """
-    Returns the client platform based on the system type.
-    """
-    if config.client.platform == "auto":
-        system_type: str = platform.system()
-
-        # macOS
-        if system_type == "Darwin":
-            config.client.platform = "mac"
-
-        # Linux
-        elif system_type == "Linux":
-            try:
-                with open("/proc/device-tree/model", "r", encoding="utf-8") as m:
-                    if "raspberry pi" in m.read().lower():
-                        config.client.platform = "rpi"
-                    else:
-                        config.client.platform = "linux"
-            except FileNotFoundError:
-                config.client.platform = "linux"
-
-
-def handle_exit(signum, frame) -> NoReturn:  # pylint: disable=unused-argument
-    """
-    Handle exit signal.
-    """
-    os._exit(0)
-
-
-def start_client(config) -> Thread:
-    """
-    Start the client.
-    """
-    module = importlib.import_module(
-        f".clients.{config.client.platform}.device", package="source"
-    )
-
-    client_thread = Thread(target=module.main, args=[config.client.url])
-    client_thread.start()
-    return client_thread
-
-
-def start_server(config) -> Thread:
-    """
-    Start the server.
-    """
-    loop: asyncio.AbstractEventLoop = asyncio.new_event_loop()
-    asyncio.set_event_loop(loop)
-
-    server_thread = Thread(
-        target=loop.run_until_complete,
-        args=(
-            main(
-                config.server.host,
-                config.server.port,
-                config.llm.service,
-                config.llm.model,
-                config.llm.vision_enabled,
-                config.llm.functions_enabled,
-                config.llm.context_window,
-                config.llm.max_tokens,
-                config.llm.temperature,
-                config.tts.service,
-                config.stt.service,
-            ),
-        ),
-    )
-
-    server_thread.start()
-    return server_thread
-
-
-def start_tunnel(config) -> Thread:
-    """
-    Start the tunnel.
-    """
-    tunnel_thread = Thread(
-        target=create_tunnel,
-        args=[config.tunnel.service, config.server.host, config.server.port],
-    )
-
-    tunnel_thread.start()
-    return tunnel_thread
diff --git a/software/source/server/i.py b/software/source/server/i.py
index e14d6f9..3fe01fd 100644
--- a/software/source/server/i.py
+++ b/software/source/server/i.py
@@ -9,6 +9,8 @@ from pathlib import Path
 from interpreter import OpenInterpreter
 import shutil
 
+from source import config
+from source.core.config import APP_PREFIX
 
 system_message = r"""
 
@@ -189,7 +191,7 @@ def configure_interpreter(interpreter: OpenInterpreter):
     interpreter.llm.supports_vision = True
     interpreter.shrink_images = True  # Faster but less accurate
 
-    interpreter.llm.model = "gpt-4"
+    interpreter.llm.model = config.llm.model
 
     interpreter.llm.supports_functions = False
     interpreter.llm.context_window = 110000
diff --git a/software/source/utils/system.py b/software/source/utils/system.py
new file mode 100644
index 0000000..9ae2b78
--- /dev/null
+++ b/software/source/utils/system.py
@@ -0,0 +1,12 @@
+"""
+System utility functions
+"""
+
+import os
+
+
+def handle_exit(signum, frame) -> None:  # pylint: disable=unused-argument
+    """
+    Handle exit signal.
+    """
+    os._exit(0)
diff --git a/software/start.py b/software/start.py
index 10782ec..01175bb 100644
--- a/software/start.py
+++ b/software/start.py
@@ -2,67 +2,164 @@
 Application entry point
 """
 
+import asyncio
+import importlib
+import os
+import platform
 import signal
-from threading import Thread
+import threading
+
+import typer
 
 from source import config
-from source.core.utils import (
-    get_client_platform,
-    handle_exit,
-    start_client,
-    start_server,
-    start_tunnel,
-)
 from source.server.utils.local_mode import select_local_model
+from source.utils.system import handle_exit
+
+app = typer.Typer()
 
 
-def run() -> None:
+@app.command()
+def start(
+    ctx: typer.Context,
+    server: bool = typer.Option(False, "--server", help="Run server"),
+    server_host: str = typer.Option(
+        "0.0.0.0",
+        "--server-host",
+        help="Specify the server host where the server will deploy",
+    ),
+    server_port: int = typer.Option(
+        10001,
+        "--server-port",
+        help="Specify the server port where the server will deploy",
+    ),
+    tunnel_service: str = typer.Option(
+        "ngrok", "--tunnel-service", help="Specify the tunnel service"
+    ),
+    expose: bool = typer.Option(False, "--expose", help="Expose server to internet"),
+    client: bool = typer.Option(False, "--client", help="Run client"),
+    server_url: str = typer.Option(
+        None,
+        "--server-url",
+        help="Specify the server URL that the client should expect. Defaults to server-host and server-port",
+    ),
+    client_type: str = typer.Option(
+        "auto", "--client-type", help="Specify the client type"
+    ),
+    llm_service: str = typer.Option(
+        "litellm", "--llm-service", help="Specify the LLM service"
+    ),
+    model: str = typer.Option(None, "--model", help="Specify the model"),
+    llm_supports_vision: bool = typer.Option(
+        False,
+        "--llm-supports-vision",
+        help="Specify if the LLM service supports vision",
+    ),
+    llm_supports_functions: bool = typer.Option(
+        False,
+        "--llm-supports-functions",
+        help="Specify if the LLM service supports functions",
+    ),
+    context_window: int = typer.Option(
+        2048, "--context-window", help="Specify the context window size"
+    ),
+    max_tokens: int = typer.Option(
+        4096, "--max-tokens", help="Specify the maximum number of tokens"
+    ),
+    temperature: float = typer.Option(
+        0.8, "--temperature", help="Specify the temperature for generation"
+    ),
+    tts_service: str = typer.Option(
+        "openai", "--tts-service", help="Specify the TTS service"
+    ),
+    stt_service: str = typer.Option(
+        "openai", "--stt-service", help="Specify the STT service"
+    ),
+    local: bool = typer.Option(
+        False, "--local", help="Use recommended local services for LLM, STT, and TTS"
+    ),
+) -> None:
     """
-    Run the application.
+    Setup the application.
     """
-    # Set up signal handler for SIGINT (keyboard interrupt)
     signal.signal(signal.SIGINT, handle_exit)
+    config.apply_cli_args(ctx.params)
 
-    # If platform is set to auto, determine user's platform automatically.
-    if config.client.platform == "auto":
-        get_client_platform(config)
-
-    # If local mode is enabled, set up local services
     if config.local.enabled:
-        config.tts.service = config.local.tts_service
-        config.stt.service = config.local.stt_service
-        select_local_model()
+        config.tts.service = "piper"
+        config.stt.service = "local-whisper"
 
-    # If no client URL is provided, set one using server host and port.
-    config.client.url = (
-        config.client.url or f"{config.server.host}:{config.server.port}"
-    )
+        if not model:
+            select_local_model()
+
+    if not server_url:
+        server_url = f"{config.server.host}:{config.server.port}"
 
     if not config.server.enabled and not config.client.enabled:
-        config.server.enabled = config.client.enabled = True
+        config.server.enabled = True
+        config.client.enabled = True
 
-    server_thread: Thread | None = (
-        start_server(config) if config.server.enabled else None
-    )
+    # Temporary fix pending refactor of `server` and `tunnel` modules.
+    # Prevents early execution of top-level code until config is fully initialized.
+    server_module = importlib.import_module("source.server.server")
+    tunnel_module = importlib.import_module("source.server.tunnel")
 
-    tunnel_thread: Thread | None = (
-        start_tunnel(config) if config.tunnel.exposed else None
-    )
+    if config.server.enabled:
+        loop: asyncio.AbstractEventLoop = asyncio.new_event_loop()
+        asyncio.set_event_loop(loop)
+        server_thread = threading.Thread(
+            target=loop.run_until_complete,
+            args=(
+                server_module.main(
+                    config.server.host,
+                    config.server.port,
+                    config.llm.service,
+                    config.llm.model,
+                    config.llm.vision_enabled,
+                    config.llm.functions_enabled,
+                    config.llm.context_window,
+                    config.llm.max_tokens,
+                    config.llm.temperature,
+                    config.tts.service,
+                    config.stt.service,
+                ),
+            ),
+        )
+        server_thread.start()
 
-    client_thread: Thread | None = (
-        start_client(config) if config.client.enabled else None
-    )
+    if config.tunnel.exposed:
+        tunnel_thread = threading.Thread(
+            target=tunnel_module.create_tunnel,
+            args=[config.tunnel.service, config.server.host, config.server.port],
+        )
+        tunnel_thread.start()
+
+    if config.client.enabled:
+        if config.client.platform == "auto":
+            system: str = platform.system()
+            if system == "Darwin":  # macOS
+                config.client.platform = "mac"
+            elif system == "Linux":
+                try:
+                    with open("/proc/device-tree/model", "r", encoding="utf-8") as m:
+                        if "raspberry pi" in m.read().lower():
+                            config.client.platform = "rpi"
+                        else:
+                            config.client.platform = "linux"
+                except FileNotFoundError:
+                    config.client.platform = "linux"
+
+        module = importlib.import_module(
+            f".clients.{config.client.platform}.device", package="source"
+        )
+        client_thread = threading.Thread(target=module.main, args=[server_url])
+        client_thread.start()
 
     try:
-        if server_thread:
+        if config.server.enabled:
             server_thread.join()
-        if tunnel_thread:
+        if config.tunnel.exposed:
             tunnel_thread.join()
-        if client_thread and client_thread.is_alive():
+        if config.client.enabled:
             client_thread.join()
     except KeyboardInterrupt:
-        handle_exit(signal.SIGINT, None)
-
-
-if __name__ == "__main__":
-    run()
+        os.kill(os.getpid(), signal.SIGINT)