🔭 Galileo

What 🔭 Galileo does

Galileo is an AI evaluation, observability, and real-time protection platform built for GenAI applications, AI agents, and RAG workflows. The company targets AI/ML engineering teams and enterprise AI builders that need to ship LLM-powered systems into production with measurable reliability, and it is used by named customers spanning enterprise technology (Outshift by Cisco, MongoDB, HP), AI-native companies (Writer, Ema), and financial technology (Clearwater Analytics, Satisfi Labs). Its stated purpose is to bridge the gap between offline AI testing and online production safety, addressing a core blocker in enterprise GenAI deployment.

The platform is organized around the eval-to-guardrail lifecycle. Core modules include Agent Reliability for end-to-end agent measurement, Signals for observability and insights across traces and structured data, Protect for production governance and guardrails, and Eval Engineering for dataset construction and metric tuning. The proprietary Luna-2 family of small language models, launched in 2025, distills expensive LLM-as-judge evaluators into compact models that run at low latency on L4 GPUs and monitor 100% of production traffic at a claimed 96% lower cost than traditional approaches. The platform offers 20+ out-of-box evaluations covering RAG, agents, safety, and security, with custom evaluators for domain-specific use cases, and supports text, structured data, code, and multimodal signals. Strategic integrations include NVIDIA NeMo and NIM (with NeMo Guardrails incorporated into Galileo Protect), HP AI Studio, and CrewAI.

Galileo operates a hybrid go-to-market: a self-serve SaaS model with a free tier for developers, paired with an enterprise sales-assisted motion offering Virtual Private Cloud and On-Premises deployment for buyers with data sovereignty and compliance requirements. Pricing is not publicly disclosed. The company pursues a content-led developer marketing strategy spanning LinkedIn, YouTube, GitHub, X, blogs, ebooks, podcasts, research publications, and a public Agent Leaderboard hosted on Hugging Face. Revenue mechanics, headcount, and funding history are not disclosed in the available source material.

Name

🔭 Galileo

Website

https://rungalileo.io

Company type

Private

Founded year

2021

Operating status

Operating

Short description

Galileo provides an AI evaluation, observability, and real-time protection platform for GenAI applications, agents, and RAG workflows, serving enterprise AI teams and developers with proprietary Luna-2 small language models and an eval-to-guardrail lifecycle.

Ownership category

akta.pro rank

Product category

AI Observability Platform

NAICS

Navigational, Measuring, Electromedical, and Control Instruments Manufacturing (33451)

SIC

Services-Testing Laboratories (8734)

akta.pro primary industry

AI Observability, Monitoring & Evaluation Platforms (Drift, Quality, Safety) (HDAEANAF)

akta.pro secondary industry

Safety & Alignment Evaluation (red-teaming, harmful capability testing) (HDAAAMAL)

Keywords

• AI observability platform
• LLM evaluation tools
• AI agent monitoring
• RAG evaluation
• AI guardrails

Headquarters

HQ city

South San Francisco

HQ country

United States

HQ region

North America

Markets served

GTM type

B2B

Offering type

Software

Cost components

Technology or R&D, Personnel, Marketing or Sales, Infrastructure, Operations

Go-to-market motion1 record

Distribution channels4 records

Marketing channels10 records

Core offering

Galileo provides an AI observability and eval engineering platform that converts offline evaluations into production guardrails for GenAI applications, AI agents, and RAG workflows. It ingests signals (models, prompts, functions, context, datasets, traces), surfaces failure modes via an insights engine, and enforces low-latency guardrails through proprietary Luna-2 small language models distilled from LLM-as-judge evaluators.

Product overview

Galileo AI is an AI observability and eval engineering platform organized as a unified platform with multiple integrated modules. The core platform includes Agent Reliability for end-to-end agent measurement, Signals for observability and insights ingestion, Luna-2 small language models for cost-efficient evaluation at scale, and Protect for production guardrails. The Eval Engineering module provides methodology support for advanced workflows. Products work together to enable the eval-to-guardrail lifecycle where pre-production evaluations become production governance. The platform supports SaaS, Virtual Private Cloud, and On-Premises deployment options.

Differentiator

Problem solved

Functional benefit

Quantifiable outcome

• 96% lower cost to monitor 100% of production traffic
• +2 more outcomes

Named customers6 records

Segments1 record

Ideal customer profiles2 records

Technology focussed Yes

API detail

Has API

Yes

API docs

API detail

Core technology

AI maturity

App detail

Integration3 records

AI capability9 records

Feature5 records

Partnerships

Four partnerships are on record, tiered core.

• NVIDIAcoreTechnology or IntegrationIntegration with NVIDIA NeMo for AI data flywheel capabilities and NVIDIA NIM for production monitoring. Galileo Protect incorporates NeMo Guardrails. Real-time observability evaluates systems running on NVIDIA NIM, sending alerts for drift or issues.
• HPcoreTechnology or IntegrationHP AI Studio integrated with Galileo's technology to enable data scientists to detect and correct hallucinations, drift, and bias. Galileo provides the trust framework capabilities within HP's AI Studio platform.
• CrewAIcoreTechnology or IntegrationStrategic partnership enabling CrewAI customers to use Galileo's evaluation technology. CrewAI CEO states partnership helps companies 'move fast and stay reliable' with agents that work at scale in production.
• Cisco (Outshift)coreTechnology or IntegrationOutshift by Cisco, Cisco's incubation arm for AI, has partnered with Galileo. Cisco SVP states Galileo provides tools to 'measure agent behavior, optimize performance, and ensure reliable operations' for faster production deployment.

Recent moves6 records

Expansion highlights5 records

Direct peers

• LangSmith (LangChain): LangSmith provides debugging, testing, evaluation, and monitoring for LLM applications built with LangChain. It overlaps directly with Galileo's evaluation and observability capabilities for GenAI and agent workflows.
• Arize AI: Arize AI is a leading AI observability and evaluation platform for LLMs and agents, offering tracing, evaluation, drift detection, and guardrails. It is the closest direct competitor to Galileo in enterprise AI reliability tooling.
• WhyLabs: WhyLabs offers an AI observability platform with model monitoring, data quality, and LLM evaluation capabilities. It competes with Galileo in production-time monitoring of GenAI and traditional ML systems.
• Arthur AI: Arthur AI delivers an AI observability and governance platform covering LLM performance, evaluation, and guardrails for production deployments. It is a direct functional competitor to Galileo's agent reliability and protect offerings.
• Fiddler AI: Fiddler AI provides an AI observability and explainability platform for LLM and ML model monitoring, performance, and responsible AI. Its scope on production monitoring and guardrails overlaps materially with Galileo's evaluation and Protect modules.

Emerging players

• Helicone: Helicone is an emerging LLM observability platform providing logging, tracing, and evaluation for AI applications. It targets overlapping developer personas but with a narrower scope than Galileo's full eval-to-guardrail lifecycle.
• Confident AI (DeepEval): Confident AI offers DeepEval, an open-source LLM evaluation framework, and a cloud platform for LLM testing. It competes with Galileo in the evaluation-engine category, particularly with developer-led adoption.
• Patronus AI: Patronus AI focuses on automated evaluation and safety testing for LLM applications, including hallucination and safety detection. It overlaps with Galileo's safety/security evals and is a comparable emerging player in AI evaluation.

Broad incumbents

• Datadog: Datadog is a broad observability and monitoring platform that has added LLM/AI monitoring capabilities (e.g., LLM Observability, watchdogs for AI applications). It competes indirectly by bundling AI evaluation into a larger observability suite.
• Honeycomb: Honeycomb provides production observability for modern software systems and is extending into AI/LLM tracing and evaluation. It is an incumbent whose bundled approach could pressure Galileo on enterprise procurement decisions.

Market position

Strengths4 records

Weaknesses4 records

Competitive moat5 records

Key risks6 records

Key highlights6 records

Customer concentration

• Xx.com/rungalileo
• GitHubgithub.com/rungalileo
• YouTubeyoutube.com/@rungalileo
• LinkedInlinkedin.com/company/run-galileo

Revenue estimate

Valuation estimate

Number of profiles

Funding overview

Funding rounds

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about 🔭 Galileo