databufflabs/databuff

DataBuff is an AI-native APM built on Opentelemetry，with multi-agent troubleshooting out of the box.

What it solves

DataBuff is an AI-native Application Performance Monitoring (APM) backend that simplifies the process of troubleshooting distributed systems. It replaces the need for manual query language expertise by allowing users to identify root causes of performance bottlenecks and service failures using natural language queries and AI agents.

How it works

Built on OpenTelemetry (OTLP), DataBuff ingests traces, metrics, and logs. It uses an "AI Brain" to orchestrate a team of specialized AI agents (query, inspection, ops, and Q&A experts) that query live telemetry data directly to perform root-cause analysis and generate incident reports. The system supports OTLP-native ingestion, eBPF for non-intrusive collection, and is compatible with SkyWalking.

Who it’s for

It is designed for DevOps and SRE teams who need a production-grade, self-hosted observability platform that can automate the troubleshooting loop from detection to remediation.

Highlights

  • Multi-Agent Troubleshooting: Orchestrates multiple AI experts in parallel to diagnose issues and synthesize reports.
  • OTLP-Native: Full support for OpenTelemetry gRPC and HTTP ingestion.
  • Natural Language Querying: Allows users to ask questions like "which service was slowest" without writing complex queries.
  • eBPF Integration: Provides kernel-level, non-intrusive data collection without requiring code changes.
  • Flexible Model Support: Compatible with various LLMs including DeepSeek, Kimi, GLM, and Ollama.

Related

  • Project
    traceloop/openllmetry

    An open-source observability framework built on OpenTelemetry that provides complete tracing and monitoring for LLM applications, providers, and vector databases.

  • Project
    qinshihu/itops-agent-platform

    An enterprise-grade AIOps platform that uses multi-agent AI collaboration to automate the full cycle of alert diagnosis, human-approved repair, and verification.

  • Project
    openlit/openlit

    An open-source observability and evaluation platform for AI agents that uses OpenTelemetry to trace, debug, and optimize LLM applications.

  • Project
    mlflow/mlflow

    An open-source AI engineering platform that enables teams to debug, evaluate, monitor, and optimize LLMs, AI agents, and ML models for production.

  • Project
    langwatch/langwatch

    An open-source platform for AI production that allows companies to trace, test, route, and govern all LLM calls and coding agents.