diff --git a/packages/opentelemetry-instrumentation-alephalpha/README.md b/packages/opentelemetry-instrumentation-alephalpha/README.md
index 34ff056880..016b87e394 100644
--- a/packages/opentelemetry-instrumentation-alephalpha/README.md
+++ b/packages/opentelemetry-instrumentation-alephalpha/README.md
@@ -20,6 +20,79 @@ from opentelemetry.instrumentation.alephalpha import AlephAlphaInstrumentor
 AlephAlphaInstrumentor().instrument()
 ```
 
+## Beginner example
+
+The following example makes one completion request and prints the trace to the
+terminal. It is useful for understanding the complete flow locally before
+connecting the application to an observability backend.
+
+Install the instrumentation package and the Aleph Alpha client:
+
+```bash
+pip install opentelemetry-instrumentation-alephalpha aleph-alpha-client opentelemetry-sdk
+```
+
+Set your Aleph Alpha API token:
+
+```bash
+export AA_TOKEN="your-api-token"
+```
+
+Create `trace_completion.py`:
+
+```python
+import os
+
+from aleph_alpha_client import Client, CompletionRequest, Prompt
+from opentelemetry import trace
+from opentelemetry.sdk.trace import TracerProvider
+from opentelemetry.sdk.trace.export import (
+    SimpleSpanProcessor,
+    ConsoleSpanExporter,
+)
+from opentelemetry.instrumentation.alephalpha import AlephAlphaInstrumentor
+
+
+def main():
+    tracer_provider = TracerProvider()
+    tracer_provider.add_span_processor(SimpleSpanProcessor(ConsoleSpanExporter()))
+    trace.set_tracer_provider(tracer_provider)
+
+    AlephAlphaInstrumentor().instrument(tracer_provider=tracer_provider)
+
+    client = Client(token=os.environ["AA_TOKEN"])
+    request = CompletionRequest(
+        prompt=Prompt.from_text("Explain ETL in one sentence."),
+        maximum_tokens=100,
+    )
+    response = client.complete(request, model="luminous-base")
+
+    print(response.completions[0].completion)
+
+
+if __name__ == "__main__":
+    main()
+```
+
+Run it with:
+
+```bash
+python trace_completion.py
+```
+
+The terminal prints the OpenTelemetry span before the model response because
+the span is exported when `client.complete()` finishes. The span contains
+information such as the model name, request type, duration, and token usage.
+Prompt and completion content are also recorded by default; see the
+[Privacy](#privacy) section if your application handles sensitive data.
+
+The example demonstrates the four steps involved in tracing an LLM call:
+
+1. Create an OpenTelemetry tracer provider.
+2. Add an exporter that prints completed spans.
+3. Instrument the Aleph Alpha client.
+4. Make a normal client request. The instrumentation creates the span.
+
 ## Privacy
 
 **By default, this instrumentation logs prompts, completions, and embeddings to span attributes**. This gives you a clear visibility into how your LLM application is working, and can make it easy to debug and evaluate the quality of the outputs.
diff --git a/packages/opentelemetry-instrumentation-alephalpha/examples/trace_completion.py b/packages/opentelemetry-instrumentation-alephalpha/examples/trace_completion.py
new file mode 100644
index 0000000000..ca38d36704
--- /dev/null
+++ b/packages/opentelemetry-instrumentation-alephalpha/examples/trace_completion.py
@@ -0,0 +1,32 @@
+"""Trace one Aleph Alpha completion and print the span locally."""
+
+import os
+
+from aleph_alpha_client import Client, CompletionRequest, Prompt
+from opentelemetry import trace
+from opentelemetry.instrumentation.alephalpha import AlephAlphaInstrumentor
+from opentelemetry.sdk.trace import TracerProvider
+from opentelemetry.sdk.trace.export import ConsoleSpanExporter, SimpleSpanProcessor
+
+
+def main() -> None:
+    """Make one traced completion request and print the response."""
+
+    tracer_provider = TracerProvider()
+    tracer_provider.add_span_processor(SimpleSpanProcessor(ConsoleSpanExporter()))
+    trace.set_tracer_provider(tracer_provider)
+
+    AlephAlphaInstrumentor().instrument(tracer_provider=tracer_provider)
+
+    client = Client(token=os.environ["AA_TOKEN"])
+    request = CompletionRequest(
+        prompt=Prompt.from_text("Explain ETL in one sentence."),
+        maximum_tokens=100,
+    )
+    response = client.complete(request, model="luminous-base")
+
+    print(response.completions[0].completion)
+
+
+if __name__ == "__main__":
+    main()