Documentation Index
Fetch the complete documentation index at: https://docs-v1.latitude.so/llms.txt
Use this file to discover all available pages before exploring further.
Overview
This guide shows you how to integrate Latitude Telemetry into an existing application that uses the official LlamaIndex SDK.
After completing these steps:
- Every LlamaIndex call (e.g.
query) can be captured as a log in Latitude.
- Logs are grouped under a prompt, identified by a
path, inside a Latitude project.
- You can inspect inputs/outputs, measure latency, and debug LlamaIndex-powered features from the Latitude dashboard.
You’ll keep calling LlamaIndex exactly as you do today — Telemetry simply
observes and enriches those calls.
Requirements
Before you start, make sure you have:
- A Latitude account and API key
- A Latitude project ID
- A Node.js or Python-based project that uses the LlamaIndex SDK
That’s it — prompts do not need to be created ahead of time.
Steps
Install requirements
Add the Latitude Telemetry package to your project:npm add @latitude-data/telemetry
pip install latitude-telemetry
Wrap your LlamaIndex-powered feature
Initialize Latitude Telemetry and wrap the code that calls LlamaIndex using telemetry.capture.import { LatitudeTelemetry } from '@latitude-data/telemetry'
import * as LlamaIndex from 'llamaindex'
import { Settings } from 'llamaindex'
import { openai } from '@llamaindex/openai'
import { agent } from '@llamaindex/workflow'
const telemetry = new LatitudeTelemetry(
process.env.LATITUDE_API_KEY,
{ instrumentations: { llamaindex: LlamaIndex } }
)
async function generateSupportReply(input: string) {
return telemetry.capture(
{
projectId: 123, // The ID of your project in Latitude
path: 'generate-support-reply', // Add a path to identify this prompt in Latitude
},
async () => {
Settings.llm = openai({ model: 'gpt-4o' })
const myAgent = agent({ tools: [] })
const response = await myAgent.run(input)
return response
}
)
}
You can use the capture method as a decorator (recommended) or as a context manager:Using decorator (recommended)
import os
from llama_index.core import VectorStoreIndex, SimpleDirectoryReader
from latitude_telemetry import Telemetry, Instrumentors, TelemetryOptions
telemetry = Telemetry(
os.environ["LATITUDE_API_KEY"],
TelemetryOptions(instrumentors=[Instrumentors.LlamaIndex]),
)
@telemetry.capture(
project_id=123, # The ID of your project in Latitude
path="generate-support-reply", # Add a path to identify this prompt in Latitude
)
def generate_support_reply(input: str) -> str:
documents = SimpleDirectoryReader("data").load_data()
index = VectorStoreIndex.from_documents(documents)
query_engine = index.as_query_engine()
response = query_engine.query(input)
return str(response)
import os
from llama_index.core import VectorStoreIndex, SimpleDirectoryReader
from latitude_telemetry import Telemetry, Instrumentors, TelemetryOptions
telemetry = Telemetry(
os.environ["LATITUDE_API_KEY"],
TelemetryOptions(instrumentors=[Instrumentors.LlamaIndex]),
)
def generate_support_reply(input: str) -> str:
with telemetry.capture(
project_id=123, # The ID of your project in Latitude
path="generate-support-reply", # Add a path to identify this prompt in Latitude
):
documents = SimpleDirectoryReader("data").load_data()
index = VectorStoreIndex.from_documents(documents)
query_engine = index.as_query_engine()
response = query_engine.query(input)
return str(response)
The path:
- Identifies the prompt in Latitude
- Can be new or existing
- Should not contain spaces or special characters (use letters, numbers,
- _ / .)
Seeing your logs in Latitude
Once your feature is wrapped, logs will appear automatically.
- Open the prompt in your Latitude dashboard (identified by
path)
- Go to the Traces section
- Each execution will show:
- Input and output messages
- Model and token usage
- Latency and errors
- One trace per feature invocation
Each LlamaIndex call appears as a child span under the captured prompt execution, giving you a full, end-to-end view of what happened.
That’s it
No changes to your LlamaIndex calls, no special return values, and no extra plumbing — just wrap the feature you want to observe.
