Liudmila Molkova lmolkova

Title	Link	# of reactions	Reactions
Capture request and response bodies	open-telemetry/semantic-conventions#857	29	{'+1': 28, '-1': 1, 'laugh': 0, 'hooray': 0, 'confused': 0, 'heart': 0, 'rocket': 0, 'eyes': 0}
Introduce semantic conventions for modern AI (LLMs, vector databases, etc.)	open-telemetry/semantic-conventions#327	11	{'+1': 0, '-1': 0, 'laugh': 0, 'hooray': 8, 'confused': 0, 'heart': 3, 'rocket': 0, 'eyes': 0}
Improve GraphQL semantic conventions	open-telemetry/semantic-conventions#182	8	{'+1': 1, '-1': 0, 'laugh': 0, 'hooray': 0, 'confused': 0, 'heart': 0, 'rocket': 6, 'eyes': 1}
Discuss ECS' field reuse concept for OTel semantic attributes	open-telemetry/semantic-conventions#339	7	{'+1': 7, '-1': 0, 'laugh': 0, 'hooray': 0, 'confused': 0, 'heart': 0, 'rocket': 0, 'eyes': 0}
Define semantic conven

Prereqs

Which one to use Otel or Micrometer. Why?
Semantic conventions - how users can pick which ones they want? will libs need to provide all possible options?
What's the story in other languages/ecosystems? I need my python messaging consumer to be consistent with my java messaging producer.
Which libraries beyond Spring are all-in on Micrometer?
How do you prioritize features? What if something is missing? (E.g. structured events on spans, adding link after span creation, etc)
- is Micrometer targeting feature-parity with OTel specification? Will micrometer introduce profling API?
How big is the community? Are there community calls? Who makes decisions?

Tested processor performance in the following setup:

Sender sends up to 4k messages per second in batches. Message size is 12 bytes.

	private string GetFinishReason(ChatFinishReason finishReason) => finishReason switch
	{
	ChatFinishReason.Stop => "stop",
	ChatFinishReason.Length => "length",
	ChatFinishReason.ToolCalls => "tool_calls",
	// ...
	_ => finishReason.ToString(),
	};

	private void LogChoice(ChatCompletion message)

	import json
	import os
	import time
	from typing import override
	from opentelemetry.instrumentation.httpx import HTTPXClientInstrumentor
	#HTTPXClientInstrumentor().instrument()

	from openai import AssistantEventHandler
	from openai.types.beta.threads.runs import RunStep

	public class Main {

	public static void main(String[] args) throws IOException {
	var logProvider = SdkLoggerProvider.builder()
	.addLogRecordProcessor(SimpleLogRecordProcessor.create(OtlpGrpcLogRecordExporter.getDefault())).build();
	EventLoggerProvider eventLoggerProvider = SdkEventLoggerProvider.create(logProvider);

	var logger = eventLoggerProvider.get("test-logger");
	var event = new GenAIChoice()
	.setId(1)

	let get_role = (event_name:string) {
	iff(event_name == "gen_ai.choice", "assistant", split(event_name, ".")[1])
	};
	let gen_ai_spans = dependencies
	\| where isnotnull(customDimensions["gen_ai.system"])
	\| project
	operation_ParentId,
	operation_Id,
	id,
	duration,

	# set OPENAI_API_KEY
	# install otel-api and otel-sdk from main
	# install opentelemetry-instrumentation-openai from https://github.com/lmolkova/opentelemetry-python-contrib/tree/openai-events

	from time import time_ns
	from openai import OpenAI

	from typing import Optional
	from opentelemetry import trace, _logs, _events
	import opentelemetry.sdk