Skip to content
OpenObserve
Get Demo Star
OpenObserve
Get Demo Star
Open Menu Close Menu

Mixpanel → OpenObserve

Record LLM usage events in Mixpanel for product analytics while simultaneously exporting structured OTel traces to OpenObserve for observability. Mixpanel captures aggregated LLM usage per user or feature; OpenObserve stores the full span tree with token counts and latency.

Prerequisites

  • Python 3.8+
  • An OpenObserve account (cloud or self-hosted)
  • Your OpenObserve organisation ID and Base64-encoded auth token
  • A Mixpanel project token
  • An OpenAI API key (or another LLM provider)

Installation

pip install openobserve-telemetry-sdk openinference-instrumentation-openai mixpanel openai python-dotenv

Configuration

Create a .env file in your project root:

OPENOBSERVE_URL=https://api.openobserve.ai/
OPENOBSERVE_ORG=your_org_id
OPENOBSERVE_AUTH_TOKEN=Basic <your_base64_token>
OPENAI_API_KEY=your-openai-api-key
MIXPANEL_TOKEN=your-mixpanel-project-token

Instrumentation

Call OpenAIInstrumentor().instrument() and openobserve_init() before creating any clients. Track each LLM call in both Mixpanel and OpenObserve.

from dotenv import load_dotenv
load_dotenv()

from openinference.instrumentation.openai import OpenAIInstrumentor
from openobserve import openobserve_init

OpenAIInstrumentor().instrument()
openobserve_init()

from opentelemetry import trace
import os
import uuid
import mixpanel
from openai import OpenAI

tracer = trace.get_tracer(__name__)
mp = mixpanel.Mixpanel(os.environ["MIXPANEL_TOKEN"])
client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

def generate(prompt: str, user_id: str = None) -> str:
    with tracer.start_as_current_span("mixpanel.llm_call") as span:
        span.set_attribute("mixpanel.model", "gpt-4o-mini")
        span.set_attribute("mixpanel.prompt", prompt[:200])
        response = client.chat.completions.create(
            model="gpt-4o-mini",
            messages=[{"role": "user", "content": prompt}],
            max_tokens=200,
        )
        reply = response.choices[0].message.content
        trace_id = hex(span.get_span_context().trace_id)

        mp.track(user_id or str(uuid.uuid4()), "LLM Call", {
            "model": "gpt-4o-mini",
            "prompt_tokens": response.usage.prompt_tokens,
            "completion_tokens": response.usage.completion_tokens,
            "trace_id": trace_id,
        })
        return reply

result = generate("Explain distributed tracing in one sentence.", user_id="user-123")
print(result)

What Gets Captured

In OpenObserve (OTel traces):

Attribute Description
mixpanel_model The model used
mixpanel_prompt The input prompt
mixpanel_completion_length Character length of the model response
llm_token_count_prompt Prompt tokens (from OpenAI instrumentor child span)
llm_token_count_completion Completion tokens (from OpenAI instrumentor child span)
duration Request latency
span_status OK or error status

In Mixpanel (LLM Call event):

Property Description
model LLM model name
prompt_tokens Input token count
completion_tokens Output token count
trace_id OpenObserve trace ID for cross-referencing

Viewing Traces

  1. Log in to OpenObserve and navigate to Traces
  2. Filter by span name mixpanel.llm_call to see all LLM calls
  3. Use the trace_id property in Mixpanel to locate the full OTel trace in OpenObserve

Mixpanel LLM trace in OpenObserve

Next Steps

With Mixpanel and OpenObserve both instrumented, you get product analytics in Mixpanel and deep observability in OpenObserve. Use Mixpanel funnels to track feature adoption and OpenObserve dashboards to monitor LLM latency.

Read More