diff --git a/.github/vale/dicts/aiven.dic b/.github/vale/dicts/aiven.dic
index db9608af66..550610ae5a 100644
--- a/.github/vale/dicts/aiven.dic
+++ b/.github/vale/dicts/aiven.dic
@@ -42,6 +42,7 @@ datasource
 Datastax
 datastore
 dashboards
+deallocation
 Debezium
 deserialize
 Dev
@@ -60,6 +61,7 @@ failover
 fileset
 filesets
 Flink
+Flink's
 Gantt
 geocoder
 geocoding
@@ -92,6 +94,7 @@ io
 IPsec
 Java
 JMX
+JobManager
 Jolokia
 Kaggle/M
 Kafdrop
@@ -210,6 +213,7 @@ Transport Layer Security
 truststore
 truststores
 Uber
+UDFs
 uncomment
 unencrypted
 untrusted
diff --git a/_toc.yml b/_toc.yml
index 5eb6edde1c..be13f72eac 100644
--- a/_toc.yml
+++ b/_toc.yml
@@ -597,12 +597,12 @@ entries:
       - file: docs/products/flink/list-overview
         title: Overview
         entries:
+          - file: docs/products/flink/concepts/flink-architecture
+            title: Architecture overview
           - file: docs/products/flink/concepts/flink-features
             title: Aiven for Apache Flink features
           - file: docs/products/flink/concepts/managed-service-features
             title: Managed service features
-          - file: docs/products/flink/concepts/flink-architecture
-            title: Architecture overview
           - file: docs/products/flink/reference/plans-pricing
             title: Plans and pricing
           - file: docs/products/flink/reference/flink-limitations
diff --git a/docs/products/flink/concepts/flink-architecture.rst b/docs/products/flink/concepts/flink-architecture.rst
index fd0e61a876..59bcf3c6a4 100644
--- a/docs/products/flink/concepts/flink-architecture.rst
+++ b/docs/products/flink/concepts/flink-architecture.rst
@@ -1,7 +1,9 @@
 Aiven for Apache Flink® architecture
 ====================================
 
-Apache Flink is an open-source distributed stream processing framework that allows you to process data in real-time. At a high level, it has a runtime architecture consisting of two types of processes: a **JobManager** and one or more **TaskManager**.
+Aiven for Apache Flink® is a fully managed service that leverages the power of the open-source Apache Flink framework to provide distributed, stateful stream processing capabilities, allowing users to perform real-time computation with SQL efficiently.
+
+ At a high level, it has a runtime architecture consisting of two types of processes: a **JobManager** and one or more **TaskManager**.
 
 JobManager
 -----------
@@ -18,8 +20,8 @@ In a high-availability setup, there may be multiple JobManagers running in the c
 
 The JobManager in Apache Flink consists of three main components: **ResourceManager**, **Dispatcher**, and **JobMaster**. The ResourceManager is responsible for managing the allocation and deallocation of resources in the Flink cluster. Additionally, ResourceMaanger is responsible for managing **Task slots** - the unit of resource scheduling in a Flink cluster.
 
-TaskManger
-----------
+TaskManager
+-----------
 TaskManager is responsible for executing the tasks assigned to them by the JobManager and exchanging data with other TaskManagers as needed. This direct communication between TaskManagers allows for efficient data exchange and helps improve the Flink runtime performance.
 TaskManagers also communicate with the JobManager to report progress and request necessary resources. This enables the JobManager to monitor the progress of tasks and to allocate resources accordingly to ensure optimal performance.
 
diff --git a/docs/products/flink/concepts/managed-service-features.rst b/docs/products/flink/concepts/managed-service-features.rst
index 2f00bf054d..9110eb7e9a 100644
--- a/docs/products/flink/concepts/managed-service-features.rst
+++ b/docs/products/flink/concepts/managed-service-features.rst
@@ -11,7 +11,7 @@ The Flink cluster executes applications in `session mode <https://nightlies.apac
 
 Scaling a Flink cluster
 -----------------------
-Each node is equipped with a TaskManager and JobManager. We recommend scaling up your cluster to add more CPU and memory for the TaskManager before attempting to scale out so you make the best use of the resources available in the nodes.
+Each node is equipped with a TaskManager and JobManager. We recommend scaling up your cluster to add more CPU and memory for the TaskManager before attempting to scale out, so you make the best use of the resources with a minimum number of nodes. 
 
 By default, each TaskManager is configured with a single slot for maximum job isolation. It is highly recommended that you modify this option to match your requirements.
 
@@ -42,4 +42,3 @@ By enabling these integrations, you can:
 - Push service metrics to M3®, InfluxDB®, or PostgreSQL® services on Aiven.
 - Create custom OpenSearch or Grafana® dashboards to monitor the service. 
 
-The platform also has alert policies configured, so you are notified via email when a key metric rises above or below a threshold set, such as low memory or high CPU consumption.
\ No newline at end of file