---
title: "Discussion on \"Building Agent Sherlock: AI-Powered Incident Investigation with SigNoz, OpenTelemetry & Multi-Agent Reasoning\" | Hashnode"
description: "Discussion on \"Building Agent Sherlock: AI-Powered Incident Investigation with SigNoz, OpenTelemetry & Multi-Agent Reasoning\". How I built an AI-powered observability and incident investigation platform using SigNoz, OpenTelemetry, LangGraph, and modern LLMs."
url: "https://hashnode.com/posts/building-agent-sherlock-ai-powered-incident-investigation-with-signoz-opentelemetry-multi-agent-reasoning/6a66184f12bb5ec4155b0247"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [Building Agent Sherlock: AI-Powered Incident Investigation with SigNoz, OpenTelemetry & Multi-Agent Reasoning](https://skfaizantech.hashnode.dev/building-agent-sherlock-ai-powered-incident-investigation-with-signoz-opentelemetry-multi-agent-reasoning?utm_source=hashnode&utm_medium=feed)

How I built an AI-powered observability and incident investigation platform using SigNoz, OpenTelemetry, LangGraph, and modern LLMs.

Jul 26 · 17 min read · skfaizantech.hashnode.dev  
Read the article

-   [#opentelemetry](https://hashnode.com/tag/opentelemetry)
-   [#signoz](https://hashnode.com/tag/signoz)
-   [#observability](https://hashnode.com/tag/observability)
-   [#artificial-intelligence](https://hashnode.com/tag/artificial-intelligence)
-   [#python](https://hashnode.com/tag/python)
-   [#wemakedevs](https://hashnode.com/tag/wemakedevs)

## Written by

[SK Faizanuddin](https://hashnode.com/@faizanshekh)

@faizanshekh · Kolkata, India

  
[View profile](https://hashnode.com/@faizanshekh)

Faizan's Engineering Blog

Hi, I'm SK Faizanuddin, a Computer Science (AI &amp; ML) graduate from Kolkata, India. I'm passionate about AI, backend engineering, observability, and open-source software. I enjoy building intelligent systems, full-stack applications, and developer tools while continuously learning and sharing my engineering journey.

## Responses 1

Markdown supported · ⌘↵ to post

[Abhishek Srivastava](https://hashnode.com/@abhicool)

[Aug 20](https://hashnode.com/posts/building-agent-sherlock-ai-powered-incident-investigation-with-signoz-opentelemetry-multi-agent-reasoning/6a66184f12bb5ec4155b0247/comment/6a86ba79d376ee78b79102e7)

The point about AI agent failure modes being non-deterministic (tool retry loops, hallucinated confidence, silently dropped context) is underappreciated — most incident-response tooling still assumes deterministic failure. It's part of why FlowTux constrains execution to a fixed, auditable command allow-list: when the agent's own reasoning can fail in weird ways, the blast radius of an action still needs a hard ceiling.
