diff --git a/infrastructure/cluster/flux-system/kustomization.yaml b/infrastructure/cluster/flux-system/kustomization.yaml
index 986fa64..f3b734f 100644
--- a/infrastructure/cluster/flux-system/kustomization.yaml
+++ b/infrastructure/cluster/flux-system/kustomization.yaml
@@ -4,3 +4,4 @@ resources:
 - gotk-components.yaml
 - gotk-sync.yaml
 - platform.yaml
+- services.yaml
diff --git a/infrastructure/cluster/flux-system/services.yaml b/infrastructure/cluster/flux-system/services.yaml
new file mode 100644
index 0000000..cb11b20
--- /dev/null
+++ b/infrastructure/cluster/flux-system/services.yaml
@@ -0,0 +1,14 @@
+apiVersion: kustomize.toolkit.fluxcd.io/v1
+kind: Kustomization
+metadata:
+  name: services
+  namespace: flux-system
+spec:
+  interval: 10m0s
+  path: ./infrastructure/services
+  prune: true
+  dependsOn:
+    - name: platform
+  sourceRef:
+    kind: GitRepository
+    name: flux-system
\ No newline at end of file
diff --git a/infrastructure/services/openai-chat-service/deployment.yaml b/infrastructure/services/openai-chat-service/deployment.yaml
new file mode 100644
index 0000000..a27a53f
--- /dev/null
+++ b/infrastructure/services/openai-chat-service/deployment.yaml
@@ -0,0 +1,55 @@
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: openai-chat-service
+  namespace: default
+  labels:
+    app: openai-chat-service
+spec:
+  replicas: 2
+  selector:
+    matchLabels:
+      app: openai-chat-service
+  template:
+    metadata:
+      labels:
+        app: openai-chat-service
+    spec:
+      containers:
+      - name: openai-chat-service
+        image: ghcr.io/lreimer/k8s-native-java-ai-openai-chat-service:main
+        imagePullPolicy: Always
+        ports:
+        - name: http
+          containerPort: 8080
+        - name: management
+          containerPort: 9000
+        resources:
+          requests:
+            memory: "128Mi"
+            cpu: "250m"
+          limits:
+            memory: "256Mi"
+            cpu: "500m"
+        probes:
+          readinessProbe:
+            httpGet:
+              path: /q/health/ready
+              port: management
+            initialDelaySeconds: 10
+            periodSeconds: 10
+          livenessProbe:
+            httpGet:
+              path: /q/health/live
+              port: management
+            initialDelaySeconds: 10
+            periodSeconds: 10
+        env:
+        - name: QUARKUS_LANGCHAIN4J_OPENAI_API_KEY
+          valueFrom:
+            secretKeyRef:
+              name: openai-api-key
+              key: OPENAI_API_KEY
+        # - name: QUARKUS_LANGCHAIN4J_OPENAI_BASE_URL
+        #   value: "http://openai-proxy-service.default.svc.cluster.local:10000"
+        
\ No newline at end of file
diff --git a/infrastructure/services/openai-chat-service/kustomization.yaml b/infrastructure/services/openai-chat-service/kustomization.yaml
new file mode 100644
index 0000000..dffc32f
--- /dev/null
+++ b/infrastructure/services/openai-chat-service/kustomization.yaml
@@ -0,0 +1,5 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+resources:
+- deployment.yaml
+- service.yaml
diff --git a/infrastructure/services/openai-chat-service/service.yaml b/infrastructure/services/openai-chat-service/service.yaml
new file mode 100644
index 0000000..fa724be
--- /dev/null
+++ b/infrastructure/services/openai-chat-service/service.yaml
@@ -0,0 +1,13 @@
+apiVersion: v1
+kind: Service
+metadata:
+  name: openai-chat-service
+  namespace: default
+spec:
+  selector:
+    app: openai-chat-service
+  ports:
+    - protocol: TCP
+      port: 8080
+      targetPort: http
+  type: LoadBalancer
\ No newline at end of file