docs: follow up to guide navigation restructure

README.md

@@ -260,7 +260,7 @@ This writes to `docs/reference/api/openapi.json`.
 
 ## Service Discovery and Messaging
 
-Dynamo uses TCP for inter-component communication. On Kubernetes, native resources ([CRDs + EndpointSlices](docs/fern/pages/developer-guide/knowledge-base/kubernetes/kubernetes-operator/service-discovery.md)) handle service discovery. External services are optional for most deployments:
+Dynamo uses TCP for inter-component communication. On Kubernetes, native resources ([CRDs + EndpointSlices](docs/fern/pages/developer-guide/knowledge-base/concepts/system-architecture/architecture.md#discovery-plane)) handle service discovery. External services are optional for most deployments:
 
 | Deployment | etcd | NATS | Notes |
 |------------|------|------|-------|

README.zh-CN.md

@@ -231,7 +231,7 @@ cargo run -p dynamo-llm --bin generate-frontend-openapi
 
 ## 服务发现与消息传递
 
-Dynamo 使用 TCP 进行组件间通信。在 Kubernetes 上，原生资源（[CRDs + EndpointSlices](docs/fern/pages/developer-guide/knowledge-base/kubernetes/kubernetes-operator/service-discovery.md)）负责服务发现。对大多数部署来说，外部服务是可选的：
+Dynamo 使用 TCP 进行组件间通信。在 Kubernetes 上，原生资源（[CRDs + EndpointSlices](docs/fern/pages/developer-guide/knowledge-base/concepts/system-architecture/architecture.md#discovery-plane)）负责服务发现。对大多数部署来说，外部服务是可选的：
 
 | 部署 | etcd | NATS | 说明 |
 |------|------|------|------|

docs/fern/index.yml

@@ -159,8 +159,6 @@ navigation:
           - section: Model Loading
             collapsed: true
             contents:
-              - page: Model Caching
-                path: pages/kubernetes/model-deployment/model-loading/model-caching.mdx
               - page: Fast Model Loading
                 path: pages/kubernetes/model-deployment/model-loading/fast-model-loading.mdx
 
@@ -170,9 +168,6 @@ navigation:
         contents:
           - page: Overview
             path: pages/kubernetes/kv-aware-routing/overview.md
-          - page: Gateway API Routing Architecture
-            slug: gateway-api-routing-architecture
-            path: pages/developer-guide/knowledge-base/kubernetes/kubernetes-operator/gateway-api-routing.md
           - page: Using the Dynamo Frontend
             path: pages/kubernetes/kv-aware-routing/dynamo-frontend.md
           - page: Using GAIE with Dynamo
@@ -187,29 +182,10 @@ navigation:
         contents:
           - page: Overview
             path: pages/kubernetes/disaggregated-serving/overview.md
-          - page: Disaggregated Communication
-            slug: disaggregated-communication
-            path: pages/developer-guide/knowledge-base/kubernetes/kubernetes-operator/disagg-communication.md
           - page: Sizing with AIConfigurator
             slug: size-with-ai-configurator
             path: pages/kubernetes/disaggregated-serving/sizing-with-aiconfigurator.mdx
 
-      - section: Multinode
-        slug: multinode
-        icon: network-wired
-        collapsed: true
-        contents:
-          - page: Multinode Deployments
-            path: pages/kubernetes/model-deployment/multinode-deployments.md
-          - page: Grove
-            path: pages/developer-guide/knowledge-base/kubernetes/multinode/grove.md
-          - page: Topology Aware Scheduling
-            path: pages/developer-guide/knowledge-base/kubernetes/multinode/topology-aware-scheduling.md
-          - page: Topology-Aware KV Transfer
-            path: pages/developer-guide/knowledge-base/kubernetes/multinode/topology-aware-kv-transfer.md
-          - page: LWS
-            path: pages/developer-guide/knowledge-base/kubernetes/multinode/lws.md
-
       - section: KV Cache Offloading
         icon: database
         collapsed: true
@@ -231,8 +207,6 @@ navigation:
             path: pages/kubernetes/fault-tolerance/request-rejection.md
           - page: Graceful Shutdown
             path: pages/kubernetes/fault-tolerance/graceful-shutdown.md
-          - page: Shadow Engine Failover
-            path: pages/developer-guide/knowledge-base/kubernetes/kubernetes-operator/shadow-engine-failover.md
 
       - section: Auto Deployment
         icon: wand-magic-sparkles
@@ -248,35 +222,10 @@ navigation:
           - page: Dynamo Planner
             path: pages/kubernetes/auto-deployment/dynamo-planner.mdx
 
-      - section: Kubernetes Operator
-        slug: kubernetes-operator
-        icon: gears
-        collapsed: true
-        contents:
-          - page: Dynamo Operator
-            path: pages/developer-guide/knowledge-base/kubernetes/kubernetes-operator/dynamo-operator.md
-          - page: Service Discovery
-            path: pages/developer-guide/knowledge-base/kubernetes/kubernetes-operator/service-discovery.md
-          - page: Webhooks
-            path: pages/developer-guide/knowledge-base/kubernetes/kubernetes-operator/webhooks.md
-          - page: Managing Models (DynamoModel)
-            path: pages/developer-guide/knowledge-base/kubernetes/kubernetes-operator/managing-models-dynamomodel.md
-          - page: Autoscaling
-            path: pages/developer-guide/knowledge-base/kubernetes/kubernetes-operator/autoscaling.md
-          - page: Rolling Update
-            path: pages/developer-guide/knowledge-base/kubernetes/kubernetes-operator/rolling-update.md
-
       - section: Operations
         icon: screwdriver-wrench
         collapsed: open-by-default
         contents:
-          - section: Cold Start Optimizations
-            collapsed: true
-            contents:
-              - page: Dynamo Snapshot
-                path: pages/developer-guide/knowledge-base/kubernetes/kubernetes-operator/snapshot.md
-              - page: Model Express
-                path: pages/developer-guide/knowledge-base/kubernetes/model-loading/modelexpress.md
           - page: Benchmarking with AIPerf
             path: pages/kubernetes/operations/benchmarking-with-aiperf.mdx
           - page: Performance Tuning

docs/fern/pages/cli/disaggregated-serving/sizing-with-aiconfigurator.mdx

@@ -88,7 +88,7 @@ for [vLLM](../../recipes/cli-templates/vllm.mdx),
 [TensorRT-LLM](../../recipes/cli-templates/tensorrt-llm.mdx).
 
 If one worker requires more GPUs than a node provides, use a
-[Kubernetes multinode deployment](../../kubernetes/model-deployment/multinode-deployments.md) instead of treating the GPU budget as one flat local device list.
+[Kubernetes multinode deployment](../../kubernetes/installation/multinode-orchestration.md) instead of treating the GPU budget as one flat local device list.
 
 </Step>
 <Step title="Validate the recommendation" id="validate-the-recommendation">

docs/fern/pages/cli/operations/lora-adapters.md

@@ -198,7 +198,7 @@ LoRA (Low-Rank Adaptation) serves specialized model variants without duplicating
     # customer-support-lora   2       2       30s
     ```
 
-    See [Managing Models with DynamoModel](../../developer-guide/knowledge-base/kubernetes/kubernetes-operator/managing-models-dynamomodel.md) for the full CRD workflow.
+    See the [DynamoModel API reference](../../reference/kubernetes-api/full-api-reference.mdx#dynamomodel) for the CRD fields.
 
     <AccordionGroup>
       <Accordion title="Load imperatively with the worker system API">

docs/fern/pages/developer-guide/advanced-customizations/developing-with-tilt.md

@@ -191,10 +191,9 @@ built-in cert controller (based on OPA cert-controller). On startup it:
 3. Injects the CA bundle into `ValidatingWebhookConfiguration` and
    `MutatingWebhookConfiguration` resources.
 
-This matches production behavior and requires no external tooling. For
-alternative certificate management (cert-manager or external certs), see the
-[webhook documentation](../knowledge-base/kubernetes/kubernetes-operator/webhooks.md) and configure via
-`helm_values` in `tilt-settings.yaml`.
+This matches production behavior and requires no external tooling. Configure alternative certificate
+management, such as cert-manager or external certificates, through `helm_values` in
+`tilt-settings.yaml`.
 
 ## Typical Workflows