Observing agents built with LangChain

Make sure that the following prerequisites are met.

• Python 3.8 or later is installed.

• An LLM API key, Anthropic, OpenAI, or any LLM provider supported by LangChain.

• Instana is configured for your application, see Getting started.

To instrument your LangChain application with Instana, complete the following steps:

1. Install the required packages for LangChain and OpenLLMetry.

   pip install langchain langchain-anthropic traceloop-sdk

2. Set your LLM provider credentials.

   export ANTHROPIC_API_KEY="<your-api-key>"

   Note:

   Replace ANTHROPIC_API_KEY with your LLM provider's environment variable if using a different provider (e.g., OPENAI_API_KEY for OpenAI).

3. Create your LangChain application.

   A LangChain workflow monitoring with OpenLLMetry is shown in the following example:

   from langchain_anthropic import ChatAnthropic
   from langchain_core.prompts import PromptTemplate
   from traceloop.sdk import Traceloop
   from traceloop.sdk.decorators import task, workflow
   
   # Initialize OpenLLMetry
   Traceloop.init(app_name="langchain_demo")
   
   @task(name="initialize_llm")
   def initialize_model():
       return ChatAnthropic(
           model="claude-3-5-sonnet-20241022",
           temperature=0.7,
           max_tokens=512
       )
   
   @task(name="create_prompt_template")
   def create_prompt():
       return PromptTemplate(
           input_variables=["input_text"],
           template="You are a helpful AI assistant. Respond to the following: {input_text}"
       )
   
   @task(name="process_llm_query")
   def process_query(prompt, llm, input_text):
       chain = prompt | llm
       return chain.invoke({"input_text": input_text})
   
   @workflow(name="langchain_conversation_workflow")
   def run_conversation():
       llm = initialize_model()
       prompt_template = create_prompt()
   
       input_text = "Explain the concept of quantum computing in simple terms."
       response = process_query(prompt_template, llm, input_text)
       print("Response:", response.content)
   
   if __name__ == "__main__":
       run_conversation()

4. Run your LangChain application.

   python langchain_app.py

   The application generates a response:

   Response: Quantum computing is a type of computing that uses quantum-mechanical phenomena such as superposition and quantum entanglement to perform calculations. Unlike classical computers that use binary digits (0s and 1s), quantum computers use quantum bits or "qubits" that can exist in multiple states simultaneously. This allows quantum computers to perform certain types of calculations much faster than classical computers.

5. View data on Instana. After running your application, data will appear in the Instana Gen AI observability dashboard:

   

   The dashboard shows:

   • Workflow execution flow

   • Individual task execution (model initialization, prompt creation, query processing)

   • LLM interactions with latency and token usage

   • Chain operations

For common issues such as traces not appearing or connection errors, see Troubleshooting.

• Explore other AI agent frameworks supported by Instana

• Learn about cost calculation for your LLM usage

• Set up alerts for your LangChain workflows