Skip to content
← Tools / Skills
ToolNebuly/10 trustVerified

Nebuly

User analytics for LLM products: what users ask, how they feel, and where the AI fails them.

Visit official site →

About

Nebuly positions itself as the user-analytics layer for LLM applications rather than another tracing debugger. Where tools like Langfuse or LangSmith help developers inspect individual traces, Nebuly aggregates production conversations to surface topics and intents, implicit feedback signals like rephrasing or frustration, sentiment, and failure patterns by subtopic. The current site frames this as the ROI platform for enterprise AI, with dashboards for adoption, churn signals, upsell signals, and AI task success measured by outcomes.

The company is led by CEO and co-founder Francesco Signorato and lists customers including Iveco Group and D-ID. Integration is a pip-installable Python SDK covering OpenAI, Azure OpenAI, Anthropic, Cohere, Hugging Face, Vertex AI, and Bedrock, with tracking keyed on a user_id per call. It is a commercial SaaS product aimed at product and business teams running chatbots, voicebots, and agents in production.

What it can do

  • Conversation analytics: topics, intents, sentiment, emotion
  • Implicit feedback detection (rephrasing, frustration signals)
  • AI adoption, churn, and upsell signal dashboards
  • Task-success measurement by outcome, not surveys
  • SDKs for OpenAI, Anthropic, Cohere, Vertex, Bedrock, and more

Setup

  1. Sign up at nebuly.com and create an API key.
  2. Install the SDK: pip install nebuly.
  3. Call nebuly.init(api_key=...) and add user_id to your model calls.
  4. Open the dashboard to explore topics, sentiment, and failure patterns.

Prerequisites

A Nebuly account and API key; an LLM app in production to instrument.

Use cases

llm-analyticsllm-observabilityuser-analyticsai-agentsenterprise-ai

Works with

Trust score — 6/10

/10 trustVerified

Scored 6/10 (verified tier, metrics-v1). Evidence: nebuly.com verified live with current enterprise positioning; Crunchbase profile active (NebulyAI Inc); CEO Francesco Signorato quoted in Data Reply partnership announcement; customers Iveco Group and D-ID named. Strengths: differentiated user-analytics angle vs trace debuggers, named enterprise customers. Docked for: (1) no public pricing or adoption metrics; (2) proprietary SaaS with SDK only, no self-host path found. License: proprietary.

How the trust score is built

Full methodology →
Adoption · stars/downloads, log-scaled vs peers
5.0/10
Maintenance · recency, releases, issue load
6.0/10
Openness · license + docs
4.0/10
Manual review · sanity read + red-flag check
6.0/10
  • · nebuly.com verified live, now positioned as ROI platform for enterprise AI
  • · Crunchbase profile active (NebulyAI Inc)
  • · CEO Francesco Signorato quoted in Data Reply partnership announcement
  • · Named customers: Iveco Group, D-ID
  • · No public pricing or adoption metrics

Sources

Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.