Integrating Langtrace with SigNoz

Introduction

Are you concerned about the reliability and accuracy of responses from your RAG based ChatBot built using LangChain? In this post, we will explain how you can integrate Langtrace with SigNoz to ship traces from your LLM applications.

Steps

The setup is simple:

1. Install langtrace.ai's python SDK - pip install langtrace-python-sdk

2. Initialize the SDK in your code in just 1 line of code.

3. Sign up and start visualizing the traces on SigNoz

And the best part,

Both Langtrace and SigNoz are fully open source and support open telemetry out of the box. Here's a quick experiment I did for a FastAPI based RAG endpoint built using LangChain.

You can find the sample code in this link:
https://gist.github.com/karthikscale3/c13a5b7c53820ee297badd3acaa26d58

1. Sign up to SigNoz

2. Set up Langtrace SDK in your code - see line 11 (Just 1 line of code)

3. Get the SigNoz OTLP endpoint and OTLP header to start sending traces to SigNoz cloud

4. Run the server with the following code.

OTEL_RESOURCE_ATTRIBUTES=service.name=Langtrace \ OTEL_EXPORTER_OTLP_ENDPOINT="<SIGNOZE_OTLP_ENDPOINT>" \ OTEL_EXPORTER_OTLP_HEADERS="<SIGNOZ_OTLP_HEADER>" \ OTEL_EXPORTER_OTLP_PROTOCOL=grpc \ opentelemetry-instrument uvicorn main:app

That's it! ✨ 🧙‍♂️ Enjoy the high cardinality traces on SigNoz. 👇

Links

• SigNoz - https://signoz.io/

• Langtrace - https://langtrace.ai/

• Langtrace SDK - https://pypi.org/project/langtrace-python-sdk/