diff --git a/lib/core/mqtt_topics.hpp b/lib/core/mqtt_topics.hpp
index f959a29d..435095c2 100644
--- a/lib/core/mqtt_topics.hpp
+++ b/lib/core/mqtt_topics.hpp
@@ -15,7 +15,9 @@ enum class MqttTopic {
   kDisplacement,
   kVelocity,
   kAcceleration,
-  kLogs
+  kLogs,
+  kLatencyRequest,
+  kLatencyResponse
 };
 
 const std::unordered_map<MqttTopic, std::string> mqtt_topic_to_string
@@ -27,7 +29,9 @@ const std::unordered_map<MqttTopic, std::string> mqtt_topic_to_string
      {MqttTopic::kDisplacement, "hyped/cart_2024/navigation/displacement"},
      {MqttTopic::kVelocity, "hyped/cart_2024/navigation/velocity"},
      {MqttTopic::kAcceleration, "hyped/cart_2024/navigation/acceleration"},
-     {MqttTopic::kLogs, "hyped/cart_2024/logs"}};
+     {MqttTopic::kLogs, "hyped/cart_2024/logs"},
+     {MqttTopic::kLatencyRequest, "hyped/cart_2024/latency/request"},
+     {MqttTopic::kLatencyResponse, "hyped/cart_2024/latency/response"}};
 
 const std::unordered_map<std::string, MqttTopic> mqtt_string_to_topic
   = {{"hyped/cart_2024/state/state", MqttTopic::kState},
@@ -38,6 +42,8 @@ const std::unordered_map<std::string, MqttTopic> mqtt_string_to_topic
      {"hyped/cart_2024/navigation/displacement", MqttTopic::kDisplacement},
      {"hyped/cart_2024/navigation/velocity", MqttTopic::kVelocity},
      {"hyped/cart_2024/navigation/acceleration", MqttTopic::kAcceleration},
-     {"hyped/cart_2024/logs", MqttTopic::kLogs}};
+     {"hyped/cart_2024/logs", MqttTopic::kLogs},
+     {"hyped/cart_2024/latency/request", MqttTopic::kLatencyRequest},
+     {"hyped/cart_2024/latency/response", MqttTopic::kLatencyResponse}};
 
 }  // namespace hyped::core
diff --git a/lib/telemetry/latency.cpp b/lib/telemetry/latency.cpp
new file mode 100644
index 00000000..af88489a
--- /dev/null
+++ b/lib/telemetry/latency.cpp
@@ -0,0 +1,46 @@
+#include "latency.hpp"
+
+#include <core/wall_clock.hpp>
+
+namespace hyped::telemetry {
+
+Latency::Latency(std::shared_ptr<core::IMqtt> mqtt)
+{
+  mqtt_->subscribe(core::MqttTopic::kStateRequest);
+}
+
+void Latency::respond()
+{
+  const auto nextMessage = mqtt_->getMessage();
+  if (!nextMessage) { return; }
+
+  const auto payload       = nextMessage->payload;
+  const auto topic        = core::MqttTopic::kLatencyResponse;
+  const core::MqttMessage::Header header{.timestamp = 0,
+                                         .priority  = core::MqttMessagePriority::kNormal};
+  const core::MqttMessage message{topic, header, payload};
+  mqtt_->publish(message, core::MqttMessageQos::kAtLeastOnce);
+}
+
+void Latency::run()
+{
+  mqtt_->consume();
+  respond();
+}
+
+core::Result Latency::startNode(const std::string &mqtt_ip, const std::uint32_t mqtt_port)
+{
+  core::WallClock wall_clock;
+  core::Logger logger("LATENCY", core::LogLevel::kDebug, wall_clock);
+  auto optional_mqtt = core::Mqtt::create(logger, "latency", mqtt_ip, mqtt_port);
+  if (!optional_mqtt) {
+    logger.log(core::LogLevel::kFatal, "Failed to create MQTT client");
+    return core::Result::kFailure;
+  }
+  auto mqtt = *optional_mqtt;
+  Latency latency(mqtt);
+  latency.run();
+  return core::Result::kSuccess;
+}
+
+}  // namespace hyped::telemetry
diff --git a/lib/telemetry/latency.hpp b/lib/telemetry/latency.hpp
new file mode 100644
index 00000000..4a085269
--- /dev/null
+++ b/lib/telemetry/latency.hpp
@@ -0,0 +1,18 @@
+#pragma once
+
+#include <core/mqtt.hpp>
+#include <core/types.hpp>
+
+namespace hyped::telemetry {
+
+class Latency {
+ public:
+  Latency(std::shared_ptr<core::IMqtt> mqtt);
+  void respond();
+  void run();
+  static core::Result startNode(const std::string &mqtt_ip, const std::uint32_t mqtt_port);
+
+  const std::shared_ptr<core::IMqtt> mqtt_;
+};
+
+}  // namespace hyped::telemetry
diff --git a/telemetry/packages/ui/app/context/pods.tsx b/telemetry/packages/ui/app/context/pods.tsx
index 8545fa53..c462f828 100644
--- a/telemetry/packages/ui/app/context/pods.tsx
+++ b/telemetry/packages/ui/app/context/pods.tsx
@@ -152,7 +152,7 @@ export const PodsProvider = ({ children }: { children: React.ReactNode }) => {
           publish(
             'latency/request',
             JSON.stringify({
-              latency: new Date().getTime().toString(),
+              timestamp: new Date().getTime().toString(),
             }),
             podId,
           );
@@ -233,7 +233,7 @@ export const PodsProvider = ({ children }: { children: React.ReactNode }) => {
           // calculate the latency
           const latency =
             new Date().getTime() -
-            parseInt(JSON.parse(message.toString())['latency'] as string);
+            parseInt(JSON.parse(message.toString())['timestamp'] as string);
 
           // send warning to the server if the latency is too high
           if (latency > POD_WARNING_LATENCY) {