diff --git a/api/v1alpha1/k8sgpt_types.go b/api/v1alpha1/k8sgpt_types.go
index 3fc6bd8c..86fb996d 100644
--- a/api/v1alpha1/k8sgpt_types.go
+++ b/api/v1alpha1/k8sgpt_types.go
@@ -88,10 +88,12 @@ type AISpec struct {
 	BaseUrl string   `json:"baseUrl,omitempty"`
 	Region  string   `json:"region,omitempty"`
 	// +kubebuilder:default:=gpt-3.5-turbo
-	Model   string     `json:"model,omitempty"`
-	Engine  string     `json:"engine,omitempty"`
-	Secret  *SecretRef `json:"secret,omitempty"`
-	Enabled bool       `json:"enabled,omitempty"`
+	Model  string `json:"model,omitempty"`
+	Engine string `json:"engine,omitempty"`
+	// +kubebuilder:default:=0
+	Interval int        `json:"interval,omitempty"`
+	Secret   *SecretRef `json:"secret,omitempty"`
+	Enabled  bool       `json:"enabled,omitempty"`
 	// +kubebuilder:default:=true
 	Anonymize *bool `json:"anonymized,omitempty"`
 	// +kubebuilder:default:=english
diff --git a/config/crd/bases/core.k8sgpt.ai_k8sgpts.yaml b/config/crd/bases/core.k8sgpt.ai_k8sgpts.yaml
index 4a26a524..c8ab99e5 100644
--- a/config/crd/bases/core.k8sgpt.ai_k8sgpts.yaml
+++ b/config/crd/bases/core.k8sgpt.ai_k8sgpts.yaml
@@ -74,6 +74,9 @@ spec:
                     type: boolean
                   engine:
                     type: string
+                  interval:
+                    default: 0
+                    type: integer
                   language:
                     default: english
                     type: string
diff --git a/controllers/k8sgpt_controller.go b/controllers/k8sgpt_controller.go
index ae1eef2e..06687de1 100644
--- a/controllers/k8sgpt_controller.go
+++ b/controllers/k8sgpt_controller.go
@@ -34,6 +34,7 @@ import (
 	"sigs.k8s.io/controller-runtime/pkg/metrics"
 
 	kclient "github.com/k8sgpt-ai/k8sgpt-operator/pkg/client"
+	"github.com/k8sgpt-ai/k8sgpt-operator/pkg/common"
 	"github.com/k8sgpt-ai/k8sgpt-operator/pkg/integrations"
 	"github.com/k8sgpt-ai/k8sgpt-operator/pkg/resources"
 	"github.com/k8sgpt-ai/k8sgpt-operator/pkg/sinks"
@@ -77,6 +78,8 @@ var (
 	analysisRetryCount int
 	// allowBackendAIRequest a circuit breaker that switching on/off backend AI calls
 	allowBackendAIRequest = true
+	calledOnce            = false
+	latestResponse        = &common.K8sGPTReponse{}
 )
 
 // K8sGPTReconciler reconciles a K8sGPT object
@@ -88,6 +91,22 @@ type K8sGPTReconciler struct {
 	K8sGPTClient *kclient.Client
 }
 
+func repeatBackendRequest(interval time.Duration, k8sgptClient *kclient.Client, k8sgptConfig *corev1alpha1.K8sGPT) {
+	time.AfterFunc(interval, func() {
+		fmt.Println("Hello backend")
+		response, err := k8sgptClient.ProcessAnalysis(k8sgptConfig, allowBackendAIRequest)
+		if err != nil {
+			fmt.Printf("error: %s\n", err)
+			k8sgptClient.Close()
+			return
+		}
+
+		latestResponse = response
+		fmt.Println("Number of results", len(latestResponse.Results))
+		repeatBackendRequest(interval, k8sgptClient, k8sgptConfig)
+	})
+}
+
 // +kubebuilder:rbac:groups=core.k8sgpt.ai,resources=k8sgpts,verbs=get;list;watch;create;update;patch;delete
 // +kubebuilder:rbac:groups=core.k8sgpt.ai,resources=k8sgpts/status,verbs=get;update;patch
 // +kubebuilder:rbac:groups=core.k8sgpt.ai,resources=k8sgpts/finalizers,verbs=update
@@ -221,8 +240,6 @@ func (r *K8sGPTReconciler) Reconcile(ctx context.Context, req ctrl.Request) (ctr
 			return r.finishReconcile(err, false)
 		}
 
-		defer k8sgptClient.Close()
-
 		// Configure the k8sgpt deployment if required
 		if k8sgptConfig.Spec.RemoteCache != nil {
 			err = k8sgptClient.AddConfig(k8sgptConfig)
@@ -243,7 +260,7 @@ func (r *K8sGPTReconciler) Reconcile(ctx context.Context, req ctrl.Request) (ctr
 			}
 		}
 
-		response, err := k8sgptClient.ProcessAnalysis(deployment, k8sgptConfig, allowBackendAIRequest)
+		response, err := k8sgptClient.ProcessAnalysis(k8sgptConfig, allowBackendAIRequest)
 		if err != nil {
 			if k8sgptConfig.Spec.AI.Enabled {
 				k8sgptNumberOfFailedBackendAICalls.With(prometheus.Labels{
@@ -270,6 +287,18 @@ func (r *K8sGPTReconciler) Reconcile(ctx context.Context, req ctrl.Request) (ctr
 		// Reset analysisRetryCount
 		analysisRetryCount = 0
 
+		// interval := time.Duration(10) * time.Second
+		// if interval >= time.Duration(1)*time.Second && !calledOnce {
+		interval := time.Duration(k8sgptConfig.Spec.AI.Interval) * time.Second
+		if interval >= ReconcileSuccessInterval && !calledOnce {
+			calledOnce = true
+			repeatBackendRequest(interval, k8sgptClient, k8sgptConfig)
+		} else {
+			// If backend request interval is not set, close the client as soon
+			// as the reconciler call ends.
+			defer k8sgptClient.Close()
+		}
+
 		// Update metrics count
 		if k8sgptConfig.Spec.AI.Enabled && len(response.Results) > 0 {
 			k8sgptNumberOfBackendAICalls.With(prometheus.Labels{
diff --git a/pkg/client/analysis.go b/pkg/client/analysis.go
index 01816fb3..82ca58d2 100644
--- a/pkg/client/analysis.go
+++ b/pkg/client/analysis.go
@@ -9,10 +9,9 @@ import (
 	schemav1 "buf.build/gen/go/k8sgpt-ai/k8sgpt/protocolbuffers/go/schema/v1"
 	"github.com/k8sgpt-ai/k8sgpt-operator/api/v1alpha1"
 	"github.com/k8sgpt-ai/k8sgpt-operator/pkg/common"
-	v1 "k8s.io/api/apps/v1"
 )
 
-func (c *Client) ProcessAnalysis(deployment v1.Deployment, config *v1alpha1.K8sGPT, allowAIRequest bool) (*common.K8sGPTReponse, error) {
+func (c *Client) ProcessAnalysis(config *v1alpha1.K8sGPT, allowAIRequest bool) (*common.K8sGPTReponse, error) {
 
 	client := rpc.NewServerServiceClient(c.conn)
 	req := &schemav1.AnalyzeRequest{