diff --git a/mlops/unit_4_triggering/markdowns/__init__.py b/mlops/unit_4_triggering/markdowns/__init__.py
new file mode 100644
index 000000000..e69de29bb
diff --git a/mlops/unit_4_triggering/markdowns/online_inference_guide.md b/mlops/unit_4_triggering/markdowns/online_inference_guide.md
new file mode 100644
index 000000000..674738ebf
--- /dev/null
+++ b/mlops/unit_4_triggering/markdowns/online_inference_guide.md
@@ -0,0 +1,37 @@
+# Sample online inference
+
+Use the following CURL command to get real-time predictions:
+
+```curl
+curl --location 'http://localhost:6789/api/runs' \
+--header 'Authorization: Bearer fire' \
+--header 'Content-Type: application/json' \
+--header 'Cookie: lng=en' \
+--data '{
+    "run": {
+        "pipeline_uuid": "predict",
+        "block_uuid": "inference",
+        "variables": {
+            "inputs": [
+                {
+                    "DOLocationID": "239",
+                    "PULocationID": "236",
+                    "trip_distance": 1.98
+                },
+                {
+                    "DOLocationID": "170",
+                    "PULocationID": "65",
+                    "trip_distance": 6.54
+                }
+            ]
+        }
+    }
+}'
+```
+
+## Note
+
+The `Authorization` header is using this pipeline’s API trigger’s token value.
+The token value is set to `fire` for this project.
+If you create a new trigger, that token will change.
+Only use a fixed token for testing or demonstration purposes.
\ No newline at end of file
diff --git a/mlops/unit_4_triggering/pipelines/predict/metadata.yaml b/mlops/unit_4_triggering/pipelines/predict/metadata.yaml
index 1be72d5a8..d57819fa5 100755
--- a/mlops/unit_4_triggering/pipelines/predict/metadata.yaml
+++ b/mlops/unit_4_triggering/pipelines/predict/metadata.yaml
@@ -19,6 +19,23 @@ blocks:
   type: global_data_product
   upstream_blocks: []
   uuid: model
+- all_upstream_blocks_executed: true
+  color: null
+  configuration:
+    file_source:
+      path: unit_4_triggering/markdowns/online_inference_guide.md
+  downstream_blocks: []
+  executor_config: null
+  executor_type: local_python
+  has_callback: false
+  language: markdown
+  name: online inference guide
+  retry_config: null
+  status: updated
+  timeout: null
+  type: markdown
+  upstream_blocks: []
+  uuid: online_inference_guide
 - all_upstream_blocks_executed: true
   color: teal
   configuration: