diff --git a/app/javascript/dashboard/api/integrations/openapi.js b/app/javascript/dashboard/api/integrations/openapi.js
index 641cfcfbd..ad203a14c 100644
--- a/app/javascript/dashboard/api/integrations/openapi.js
+++ b/app/javascript/dashboard/api/integrations/openapi.js
@@ -2,18 +2,62 @@
 
 import ApiClient from '../ApiClient';
 
+/**
+ * Represents the data object for a OpenAI hook.
+ * @typedef {Object} ConversationMessageData
+ * @property {string} [tone] - The tone of the message.
+ * @property {string} [content] - The content of the message.
+ * @property {string} [conversation_display_id] - The display ID of the conversation (optional).
+ */
+
+/**
+ * A client for the OpenAI API.
+ * @extends ApiClient
+ */
 class OpenAIAPI extends ApiClient {
+  /**
+   * Creates a new OpenAIAPI instance.
+   */
   constructor() {
     super('integrations', { accountScoped: true });
+
+    /**
+     * The conversation events supported by the API.
+     * @type {string[]}
+     */
+    this.conversation_events = [
+      'summarize',
+      'reply_suggestion',
+      'label_suggestion',
+    ];
+
+    /**
+     * The message events supported by the API.
+     * @type {string[]}
+     */
+    this.message_events = ['rephrase'];
   }
 
+  /**
+   * Processes an event using the OpenAI API.
+   * @param {Object} options - The options for the event.
+   * @param {string} [options.type='rephrase'] - The type of event to process.
+   * @param {string} [options.content] - The content of the event.
+   * @param {string} [options.tone] - The tone of the event.
+   * @param {string} [options.conversationId] - The ID of the conversation to process the event for.
+   * @param {string} options.hookId - The ID of the hook to use for processing the event.
+   * @returns {Promise} A promise that resolves with the result of the event processing.
+   */
   processEvent({ type = 'rephrase', content, tone, conversationId, hookId }) {
+    /**
+     * @type {ConversationMessageData}
+     */
     let data = {
       tone,
       content,
     };
 
-    if (type === 'reply_suggestion' || type === 'summarize') {
+    if (this.conversation_events.includes(type)) {
       data = {
         conversation_display_id: conversationId,
       };
diff --git a/enterprise/lib/enterprise/integrations/openai_processor_service.rb b/enterprise/lib/enterprise/integrations/openai_processor_service.rb
new file mode 100644
index 000000000..3f0ee8c4c
--- /dev/null
+++ b/enterprise/lib/enterprise/integrations/openai_processor_service.rb
@@ -0,0 +1,54 @@
+module Enterprise::Integrations::OpenaiProcessorService
+  ALLOWED_EVENT_NAMES = %w[rephrase summarize reply_suggestion label_suggestion].freeze
+  CACHEABLE_EVENTS = %w[label_suggestion].freeze
+
+  def label_suggestion_message
+    payload = label_suggestion_body
+    return nil if payload.blank?
+
+    response = make_api_call(label_suggestion_body)
+
+    # LLMs are not deterministic, so this is bandaid solution
+    # To what you ask? Sometimes, the response includes
+    # "Labels:" in it's response in some format. This is a hacky way to remove it
+    # TODO: Fix with with a better prompt
+    response.gsub(/^(label|labels):/i, '')
+  end
+
+  private
+
+  def labels_with_messages
+    labels = hook.account.labels.pluck(:title).join(', ')
+
+    character_count = labels.length
+    conversation = find_conversation
+    messages = init_messages_body(false)
+    add_messages_until_token_limit(conversation, messages, false, character_count)
+
+    return nil if messages.blank? || labels.blank?
+
+    "Messages:\n#{messages}\nLabels:\n#{labels}"
+  end
+
+  def label_suggestion_body
+    content = labels_with_messages
+    return nil if content.blank?
+
+    {
+      model: self.class::GPT_MODEL,
+      messages: [
+        {
+          role: 'system',
+          content: 'Your role is as an assistant to a customer support agent. You will be provided with a transcript of a conversation between a ' \
+                   'customer and the support agent, along with a list of potential labels. ' \
+                   'Your task is to analyze the conversation and select the two labels from the given list that most accurately ' \
+                   'represent the themes or issues discussed. Ensure you preserve the exact casing of the labels as they are provided in the list. ' \
+                   'Do not create new labels; only choose from those provided. Once you have made your selections, please provide your response ' \
+                   'as a comma-separated list of the provided labels. Remember, your response should only contain the labels you\'ve selected, ' \
+                   'in their original casing, and nothing else. '
+        },
+        { role: 'user', content: content }
+      ]
+    }.to_json
+  end
+end
diff --git a/enterprise/spec/integrations/openai/processor_service_spec.rb b/enterprise/spec/integrations/openai/processor_service_spec.rb
new file mode 100644
index 000000000..6fa41f4bc
--- /dev/null
+++ b/enterprise/spec/integrations/openai/processor_service_spec.rb
@@ -0,0 +1,81 @@
+require 'rails_helper'
+
+RSpec.describe Integrations::Openai::ProcessorService do
+  subject { described_class.new(hook: hook, event: event) }
+
+  let(:account) { create(:account) }
+  let(:hook) { create(:integrations_hook, :openai, account: account) }
+  let(:expected_headers) { { 'Authorization' => "Bearer #{hook.settings['api_key']}" } }
+  let(:openai_response) do
+    {
+      'choices' => [
+        {
+          'message' => {
+            'content' => 'This is a reply from openai.'
+          }
+        }
+      ]
+    }.to_json
+  end
+  let!(:conversation) { create(:conversation, account: account) }
+  let!(:customer_message) { create(:message, account: account, conversation: conversation, message_type: :incoming, content: 'hello agent') }
+  let!(:agent_message) { create(:message, account: account, conversation: conversation, message_type: :outgoing, content: 'hello customer') }
+
+  describe '#perform' do
+    context 'when event name is label_suggestion with labels' do
+      let(:event) { { 'name' => 'label_suggestion', 'data' => { 'conversation_display_id' => conversation.display_id } } }
+      let(:label1) { create(:label, account: account) }
+      let(:label2) { create(:label, account: account) }
+      let(:label_suggestion_payload) do
+        labels = "#{label1.title}, #{label2.title}"
+        messages =
+          "Customer #{customer_message.sender.name} : #{customer_message.content}\nAgent #{agent_message.sender.name} : #{agent_message.content}"
+
+        "Messages:\n#{messages}\n\nLabels:\n#{labels}"
+      end
+
+      it 'returns the label suggestions' do
+        request_body = {
+          'model' => 'gpt-3.5-turbo',
+          'messages' => [
+            {
+              role: 'system',
+              content: 'Your role is as an assistant to a customer support agent. You will be provided with ' \
+                       'a transcript of a conversation between a customer and the support agent, along with a list of potential labels. ' \
+                       'Your task is to analyze the conversation and select the two labels from the given list that most accurately ' \
+                       'represent the themes or issues discussed. Ensure you preserve the exact casing of the labels as they are provided ' \
+                       'in the list. Do not create new labels; only choose from those provided. Once you have made your selections, ' \
+                       'please provide your response as a comma-separated list of the provided labels. Remember, your response should only contain ' \
+                       'the labels you\'ve selected, in their original casing, and nothing else. '
+            },
+            { role: 'user', content: label_suggestion_payload }
+          ]
+        }.to_json
+
+        stub_request(:post, 'https://api.openai.com/v1/chat/completions')
+          .with(body: request_body, headers: expected_headers)
+          .to_return(status: 200, body: openai_response, headers: {})
+
+        result = subject.perform
+        expect(result).to eq('This is a reply from openai.')
+      end
+    end
+
+    context 'when event name is label_suggestion with no labels' do
+      let(:event) { { 'name' => 'label_suggestion', 'data' => { 'conversation_display_id' => conversation.display_id } } }
+
+      it 'returns nil' do
+        result = subject.perform
+        expect(result).to be_nil
+      end
+    end
+
+    context 'when event name is not one that can be processed' do
+      let(:event) { { 'name' => 'unknown', 'data' => {} } }
+
+      it 'returns nil' do
+        expect(subject.perform).to be_nil
+      end
+    end
+  end
+end
diff --git a/lib/integrations/openai/processor_service.rb b/lib/integrations/openai/processor_service.rb
index 17a34e23f..61825a072 100644
--- a/lib/integrations/openai/processor_service.rb
+++ b/lib/integrations/openai/processor_service.rb
@@ -1,28 +1,18 @@
-class Integrations::Openai::ProcessorService
-  # 3.5 support 4,096 tokens
-  # 1 token is approx 4 characters
-  # 4,096 * 4 = 16,384 characters, sticking to 15,000 to be safe
-  TOKEN_LIMIT = 15_000
-  API_URL = 'https://api.openai.com/v1/chat/completions'.freeze
-  GPT_MODEL = 'gpt-3.5-turbo'.freeze
+class Integrations::Openai::ProcessorService < Integrations::OpenaiBaseService
+  def reply_suggestion_message
+    make_api_call(reply_suggestion_body)
+  end
 
-  ALLOWED_EVENT_NAMES = %w[rephrase summarize reply_suggestion].freeze
+  def summarize_message
+    make_api_call(summarize_body)
+  end
 
-  pattr_initialize [:hook!, :event!]
-
-  def perform
-    event_name = event['name']
-    return nil unless valid_event_name?(event_name)
-
-    send("#{event_name}_message")
+  def rephrase_message
+    make_api_call(rephrase_body)
   end
 
   private
 
-  def valid_event_name?(event_name)
-    ALLOWED_EVENT_NAMES.include?(event_name)
-  end
-
   def rephrase_body
     {
       model: GPT_MODEL,
@@ -42,12 +32,8 @@ class Integrations::Openai::ProcessorService
     add_messages_until_token_limit(conversation, messages, in_array_format)
   end
 
-  def find_conversation
-    hook.account.conversations.find_by(display_id: event['data']['conversation_display_id'])
-  end
-
-  def add_messages_until_token_limit(conversation, messages, in_array_format)
-    character_count = 0
+  def add_messages_until_token_limit(conversation, messages, in_array_format, start_from = 0)
+    character_count = start_from
     conversation.messages.chat.reorder('id desc').each do |message|
       character_count, message_added = add_message_if_within_limit(character_count, message, messages, in_array_format)
       break unless message_added
@@ -112,26 +98,6 @@ class Integrations::Openai::ProcessorService
       ].concat(conversation_messages(in_array_format: true))
     }.to_json
   end
-
-  def reply_suggestion_message
-    make_api_call(reply_suggestion_body)
-  end
-
-  def summarize_message
-    make_api_call(summarize_body)
-  end
-
-  def rephrase_message
-    make_api_call(rephrase_body)
-  end
-
-  def make_api_call(body)
-    headers = {
-      'Content-Type' => 'application/json',
-      'Authorization' => "Bearer #{hook.settings['api_key']}"
-    }
-
-    response = HTTParty.post(API_URL, headers: headers, body: body)
-    JSON.parse(response.body)['choices'].first['message']['content']
-  end
 end
+
+Integrations::Openai::ProcessorService.prepend_mod_with('Integrations::OpenaiProcessorService')
diff --git a/lib/integrations/openai_base_service.rb b/lib/integrations/openai_base_service.rb
new file mode 100644
index 000000000..d3f18763c
--- /dev/null
+++ b/lib/integrations/openai_base_service.rb
@@ -0,0 +1,80 @@
+class Integrations::OpenaiBaseService
+  # 3.5 support 4,096 tokens
+  # 1 token is approx 4 characters
+  # 4,096 * 4 = 16,384 characters, sticking to 15,000 to be safe
+  TOKEN_LIMIT = 15_000
+  API_URL = 'https://api.openai.com/v1/chat/completions'.freeze
+  GPT_MODEL = 'gpt-3.5-turbo'.freeze
+
+  ALLOWED_EVENT_NAMES = %w[rephrase summarize reply_suggestion].freeze
+  CACHEABLE_EVENTS = %w[].freeze
+
+  pattr_initialize [:hook!, :event!]
+
+  def perform
+    return nil unless valid_event_name?
+
+    return value_from_cache if value_from_cache.present?
+
+    response = send("#{event_name}_message")
+    save_to_cache(response) if response.present?
+
+    response
+  end
+
+  private
+
+  def event_name
+    event['name']
+  end
+
+  def cache_key
+    return nil unless event_is_cacheable?
+
+    conversation = find_conversation
+    return nil unless conversation
+
+    # since the value from cache depends on the conversation last_activity_at, it will always be fresh
+    format(::Redis::Alfred::OPENAI_CONVERSATION_KEY, event_name: event_name, conversation_id: conversation.id,
+                                                     updated_at: conversation.last_activity_at.to_i)
+  end
+
+  def value_from_cache
+    return nil unless event_is_cacheable?
+    return nil if cache_key.blank?
+
+    Redis::Alfred.get(cache_key)
+  end
+
+  def save_to_cache(response)
+    return nil unless event_is_cacheable?
+
+    Redis::Alfred.setex(cache_key, response)
+  end
+
+  def find_conversation
+    hook.account.conversations.find_by(display_id: event['data']['conversation_display_id'])
+  end
+
+  def valid_event_name?
+    # self.class::ALLOWED_EVENT_NAMES is way to access ALLOWED_EVENT_NAMES defined in the class hierarchy of the current object.
+    # This ensures that if ALLOWED_EVENT_NAMES is updated elsewhere in it's ancestors, we access the latest value.
+    self.class::ALLOWED_EVENT_NAMES.include?(event_name)
+  end
+
+  def event_is_cacheable?
+    # self.class::CACHEABLE_EVENTS is way to access CACHEABLE_EVENTS defined in the class hierarchy of the current object.
+    # This ensures that if CACHEABLE_EVENTS is updated elsewhere in it's ancestors, we access the latest value.
+    self.class::CACHEABLE_EVENTS.include?(event_name)
+  end
+
+  def make_api_call(body)
+    headers = {
+      'Content-Type' => 'application/json',
+      'Authorization' => "Bearer #{hook.settings['api_key']}"
+    }
+
+    response = HTTParty.post(API_URL, headers: headers, body: body)
+    JSON.parse(response.body)['choices'].first['message']['content']
+  end
+end
diff --git a/lib/redis/redis_keys.rb b/lib/redis/redis_keys.rb
index 0de5ff9ca..bf024b349 100644
--- a/lib/redis/redis_keys.rb
+++ b/lib/redis/redis_keys.rb
@@ -32,4 +32,5 @@ module Redis::RedisKeys
   # Check if a message create with same source-id is in progress?
   MESSAGE_SOURCE_KEY = 'MESSAGE_SOURCE_KEY::%<id>s'.freeze
   CUSTOM_FILTER_RECORDS_COUNT_KEY = 'CUSTOM_FILTER::%<account_id>d::%<user_id>d::%<filter_id>d'.freeze
+  OPENAI_CONVERSATION_KEY = 'OPEN_AI_CONVERSATION_KEY::%<event_name>s::%<conversation_id>d::%<updated_at>d'.freeze
 end
diff --git a/spec/lib/integrations/openai/processor_service_spec.rb b/spec/lib/integrations/openai/processor_service_spec.rb
index 686fa564f..6d9c414c7 100644
--- a/spec/lib/integrations/openai/processor_service_spec.rb
+++ b/spec/lib/integrations/openai/processor_service_spec.rb
@@ -97,13 +97,5 @@ RSpec.describe Integrations::Openai::ProcessorService do
         expect(result).to eq('This is a reply from openai.')
       end
     end
-
-    context 'when event name is not one that can be processed' do
-      let(:event) { { 'name' => 'unknown', 'data' => {} } }
-
-      it 'returns nil' do
-        expect(subject.perform).to be_nil
-      end
-    end
   end
 end