Users

User tracking gives you per-user visibility into token usage, costs, trace volume, and evaluation scores. This is useful for billing, abuse detection, quality monitoring, and understanding how different users interact with your application.

Adding a User ID

Set user_id via propagate_attributes() and it will automatically apply to all observations in that context:

Python SDK

from xeroml import observe, propagate_attributes

@observe()
def handle_request(query: str, user_id: str) -> str:
    with propagate_attributes(user_id=user_id):
        result = run_pipeline(query)
        return result

TypeScript SDK

import { propagateAttributes } from "@xeroml/tracing";

async function handleRequest(query: string, userId: string) {
  return await propagateAttributes({ userId }, async () => {
    return await runPipeline(query);
  });
}

OpenAI (Python)

from xeroml.openai import openai
from xeroml import propagate_attributes

def chat(message: str, user_id: str):
    with propagate_attributes(user_id=user_id):
        response = openai.chat.completions.create(
            model="gpt-4o",
            messages=[{"role": "user", "content": message}]
        )
    return response.choices[0].message.content

LangChain (Python)

from xeroml import CallbackHandler, propagate_attributes
from langchain_openai import ChatOpenAI

def chat(message: str, user_id: str):
    with propagate_attributes(user_id=user_id):
        handler = CallbackHandler()
        llm = ChatOpenAI(callbacks=[handler])
        return llm.invoke(message).content

User Dashboard

XeroML provides two views for user-level data:

User List A table of all users who have sent at least one trace. Columns include:

• Total traces
• Token usage (input / output / total)
• Estimated cost
• Average evaluation score
• Most recent trace timestamp

Individual User View Click any user to see their complete history: all traces, session timelines, feedback submitted, and aggregated metrics. Accessible at:

https://cloud.xeroml.com/project/{projectId}/users/{userId}

Custom Dashboards

For advanced user analytics, use the XeroML Metrics API to pull per-user aggregates into your own dashboards or data warehouse. You can also export user-level data via the UI export or blob storage export.