Instrumentation cookbook (#583)

* Update cookbooks

* Fix

* Fix

* Add

* Update indexg

* resolve comments

* casing

---------

Co-authored-by: Jack Gerrits <jack@jackgerrits.com>
Co-authored-by: Jack Gerrits <jackgerrits@users.noreply.github.com>

python/packages/autogen-core/docs/src/core-user-guide/cookbook/instrumenting.md

@@ -0,0 +1,35 @@
+# Instrumentating your code locally
+
+AutoGen supports instrumenting your code using [OpenTelemetry](https://opentelemetry.io). This allows you to collect traces and logs from your code and send them to a backend of your choice.
+
+While debugging, you can use a local backend such as [Aspire](https://aspiredashboard.com/) or [Jaeger](https://www.jaegertracing.io/). In this guide we will use Aspire as an example.
+
+## Setting up Aspire
+
+Follow the instructions [here](https://learn.microsoft.com/en-us/dotnet/aspire/fundamentals/dashboard/overview?tabs=bash#standalone-mode) to set up Aspire in standalone mode. This will require Docker to be installed on your machine.
+
+## Instrumenting your code
+
+Once you have a dashboard set up, now it's a matter of sending traces and logs to it. You can follow the steps in the [Telemetry Guide](../guides/telemetry.md) to set up the opentelemetry sdk and exporter.
+
+After instrumenting your code with the Aspire Dashboard running, you should see traces and logs appear in the dashboard as your code runs.
+
+## Observing LLM calls using Open AI
+
+If you are using the Open AI package, you can observe the LLM calls by setting up the opentelemetry for that library. We use [opentelemetry-instrumentation-openai](https://pypi.org/project/opentelemetry-instrumentation-openai/) in this example.
+
+Install the package:
+```bash
+pip install opentelemetry-instrumentation-openai
+```
+
+Enable the instrumentation:
+```python
+from opentelemetry.instrumentation.openai import OpenAIInstrumentor
+
+OpenAIInstrumentor().instrument()
+```
+
+Now running your code will send traces including the LLM calls to your telemetry backend (Aspire in our case).
+
+![Open AI Telemetry logs](../../images/open-ai-telemetry-example.png)

python/packages/autogen-core/docs/src/core-user-guide/guides/telemetry.md

@@ -8,6 +8,10 @@ These are the components that are currently instrumented:
 ## Instrumenting your application
 To instrument your application, you will need an sdk and an exporter. You may already have these if your application is already instrumented with open telemetry.
 
+## Clean instrumentation
+
+If you do not have open telemetry set up in your application, you can follow these steps to instrument your application.
+
 ```bash
 pip install opentelemetry-sdk
 ```
@@ -41,8 +45,25 @@ def configure_oltp_tracing(endpoint: str = None) -> trace.TracerProvider:
 
 Now you can send the trace_provider when creating your runtime:
 ```python
-single_threaded_runtime = SingleThreadedAgentRuntime(tracer_provider=provider)
-worker_runtime = WorkerAgentRuntime(tracer_provider=provider)
+# for single threaded runtime
+single_threaded_runtime = SingleThreadedAgentRuntime(tracer_provider=tracer_provider)
+# or for worker runtime
+worker_runtime = WorkerAgentRuntime(tracer_provider=tracer_provider)
 ```
 
 And that's it! Your application is now instrumented with open telemetry. You can now view your telemetry data in your telemetry backend.
+
+### Exisiting instrumentation
+
+If you have open telemetry already set up in your application, you can pass the tracer provider to the runtime when creating it:
+```python
+from opentelemetry import trace
+
+# Get the tracer provider from your application
+tracer_provider = trace.get_tracer_provider()
+
+# for single threaded runtime
+single_threaded_runtime = SingleThreadedAgentRuntime(tracer_provider=tracer_provider)
+# or for worker runtime
+worker_runtime = WorkerAgentRuntime(tracer_provider=tracer_provider)
+```

python/packages/autogen-core/docs/src/core-user-guide/index.md

@@ -94,5 +94,6 @@ cookbook/openai-assistant-agent
 cookbook/langgraph-agent
 cookbook/llamaindex-agent
 cookbook/local-llms-ollama-litellm
+cookbook/instrumenting
 
 ```