diff --git a/README.md b/README.md
index 30ad7c6..89b116e 100644
--- a/README.md
+++ b/README.md
@@ -5,7 +5,7 @@
 ## Features
 
 - Integrated web and RPC servers
-- Automated task management
+- [Automated task management with built-in start/stop controls and optional autostart](#understanding-tasks-in-pydase)
 - Event-based callback functionality for real-time updates
 - Built-in support for serving data over different protocols
 - Support for additional servers for specific use-cases
@@ -103,9 +103,48 @@ Once the server is running, you can access the web interface in a browser:
 
 In this interface, you can interact with the properties of your `Device` service.
 
+## Understanding Tasks in pydase
+
+In `pydase`, a task is defined as an asynchronous function contained in a class that inherits from `DataService`. These tasks usually contain a while loop and are designed to carry out periodic functions.
+
+For example, a task might be used to periodically read sensor data, update a database, or perform any other recurring job. The core feature of `pydase` is its ability to automatically generate start and stop functions for these tasks. This allows you to control task execution via both the frontend and an `rpyc` client, giving you flexible and powerful control over your service's operation.
+
+Another powerful feature of `pydase` is its ability to automatically start tasks upon initialization of the service. By specifying the tasks and their arguments in the `_autostart_tasks` dictionary in your service class's `__init__` method, `pydase` will automatically start these tasks when the server is started. Here's an example:
+
+```python
+from pydase import DataService, Server
+
+class SensorService(DataService):
+    def __init__(self):
+        self.readout_frequency = 1.0
+        self._autostart_tasks = {"read_sensor_data": ()}  # args passed to the function go there
+        super().__init__()
+
+    def _process_data(self, data: ...) -> None:
+        ...
+
+    def _read_from_sensor(self) -> Any:
+        ...
+
+    async def read_sensor_data(self):
+        while True:
+            data = self._read_from_sensor()
+            self._process_data(data)  # Process the data as needed
+            await asyncio.sleep(self.readout_frequency)
+
+
+if __name__ == "__main__":
+    service = SensorService()
+    Server(service).run()
+```
+
+In this example, `read_sensor_data` is a task that continuously reads data from a sensor. The readout frequency can be updated using the `readout_frequency` attribute.
+By listing it in the `_autostart_tasks` dictionary, it will automatically start running when `Server(service).run()` is executed.
+As with all tasks, `pydase` will also generate `start_read_sensor_data` and `stop_read_sensor_data` methods, which can be called to manually start and stop the data reading task.
+
 ## Documentation
 
-For more details about usage and features, see the [full documentation](URL_TO_YOUR_DOCUMENTATION).
+The full documentation provides more detailed information about `pydase`, including advanced usage examples, API references, and tips for troubleshooting common issues. See the [full documentation](URL_TO_YOUR_DOCUMENTATION) for more information.
 
 ## Contributing