diff --git a/examples/openai_client.py b/examples/openai_client.py
new file mode 100644
index 0000000..cb1ba32
--- /dev/null
+++ b/examples/openai_client.py
@@ -0,0 +1,45 @@
+import json
+import os
+from openai import OpenAI
+import dotenv
+from javelin_sdk import (
+    JavelinClient,
+    JavelinConfig,
+)
+
+javelin_api_key = os.getenv('JAVELIN_API_KEY')
+llm_api_key = os.getenv("OPENAI_API_KEY")
+
+# Create OpenAI client
+openai_client = OpenAI(api_key=llm_api_key)
+
+# Initialize Javelin Client
+config = JavelinConfig(
+    # base_url="https://api-dev.javelin.live",
+    base_url="http://localhost:8000",
+    javelin_api_key=javelin_api_key,
+)
+client = JavelinClient(config)
+client.register_openai(openai_client, route_name="openai")
+
+# Call OpenAI endpoints
+chat_completions = openai_client.chat.completions.create(
+    model="gpt-3.5-turbo",
+    messages=[{"role": "user", "content": "What is machine learning?"}],
+)
+print(chat_completions.model_dump_json(indent=2))
+
+completions = openai_client.completions.create(
+    model="gpt-3.5-turbo-instruct",
+    prompt="What is machine learning?",
+    max_tokens=7,
+    temperature=0
+)
+print(completions.model_dump_json(indent=2))
+
+embeddings = openai_client.embeddings.create(
+    model="text-embedding-ada-002",
+    input="The food was delicious and the waiter...",
+    encoding_format="float"
+)
+print(embeddings.model_dump_json(indent=2))
diff --git a/javelin_sdk/client.py b/javelin_sdk/client.py
index bb4bfc3..4d6e733 100644
--- a/javelin_sdk/client.py
+++ b/javelin_sdk/client.py
@@ -28,6 +28,7 @@ class JavelinClient:
     def __init__(self, config: JavelinConfig) -> None:
         self.config = config
         self.base_url = urljoin(config.base_url, config.api_version or "/v1")
+        
         self._headers = {
             "x-api-key": config.javelin_api_key,
         }
@@ -42,6 +43,7 @@ def __init__(self, config: JavelinConfig) -> None:
         self.bedrock_session = None
         self.default_bedrock_route = None
         self.use_default_bedrock_route = False
+        self.openai_base_url = None
 
         self.gateway_service = GatewayService(self)
         self.provider_service = ProviderService(self)
@@ -90,6 +92,116 @@ async def aclose(self):
     def close(self):
         if self._client:
             self._client.close()
+
+    def register_openai(self, 
+                    openai_client: Any, 
+                    route_name: str = None) -> Any:
+        """
+        Register the passed-in OpenAI client so that calls to:
+            - client.chat.completions.create(...)
+            - client.completions.create(...)
+            - client.embeddings.create(...)
+        are intercepted by Javelin.
+
+        Additionally sets:
+            - openai_client.base_url to self.base_url
+            - openai_client._custom_headers to include self._headers
+        """
+
+        # Store the OpenAI base URL
+        if self.openai_base_url is None:
+            self.openai_base_url = openai_client.base_url
+
+        # Point the OpenAI client to Javelin's base URL
+        openai_client.base_url=self.base_url
+
+        if not hasattr(openai_client, "_custom_headers"):
+            openai_client._custom_headers = {}
+        openai_client._custom_headers.update(self._headers)
+
+        base_url_str = str(self.openai_base_url)
+        # Remove trailing slash if present
+        if base_url_str.endswith("/"):
+            base_url_str = base_url_str[:-1]
+
+         # Update Javelin headers into the client's _custom_headers
+        openai_client._custom_headers["x-javelin-provider"] = base_url_str
+        openai_client._custom_headers["x-javelin-route"] = route_name
+
+        # Print out the headers you’ve set (for debug)
+        print("DEBUG - Patched OpenAI client headers:", openai_client._custom_headers)
+         
+        # Store references to the original methods
+        original_chat_completions_create = openai_client.chat.completions.create
+        original_completions_create = openai_client.completions.create
+        original_embeddings_create = openai_client.embeddings.create
+
+        # Define patched versions, injecting Javelin logs/traces
+
+        def patched_chat_completions_create(*args, **kwargs):
+            # BEFORE calling original
+            '''
+            TODO: self.trace_service.log_trace(
+                message="OpenAI chat.completions.create called",
+                extra={"args": args, "kwargs": kwargs},
+            )
+            '''
+
+            # Call the real method
+            response = original_chat_completions_create(*args, **kwargs)
+
+            # AFTER calling original
+            '''
+            TODO: self.trace_service.log_trace(
+                message="OpenAI chat.completions.create response",
+                extra={"response": response},
+            )
+            '''
+            return response
+
+        def patched_completions_create(*args, **kwargs):
+            '''
+            TODO: self.trace_service.log_trace(
+                message="OpenAI completions.create called",
+                extra={"args": args, "kwargs": kwargs},
+            )
+            '''
+            
+            response = original_completions_create(*args, **kwargs)
+
+            '''
+            TODO: self.trace_service.log_trace(
+                message="OpenAI completions.create response",
+                extra={"response": response},
+            )
+            '''
+            return response
+
+        def patched_embeddings_create(*args, **kwargs):
+            '''
+            TODO: self.trace_service.log_trace(
+                message="OpenAI embeddings.create called",
+                extra={"args": args, "kwargs": kwargs},
+            )
+            '''
+
+            response = original_embeddings_create(*args, **kwargs)
+
+            '''
+            TODO: self.trace_service.log_trace(
+                message="OpenAI embeddings.create response",
+                extra={"response": response},
+            )
+            '''
+            return response
+
+        # patch the client’s methods
+        openai_client.chat.completions.create = patched_chat_completions_create
+        openai_client.completions.create = patched_completions_create
+        openai_client.embeddings.create = patched_embeddings_create
+
+        # Return the patched client
+        return openai_client
             
     def register_bedrock(self, 
                     bedrock_runtime_client: Any,