Humanloop
Humanloop was an enterprise SaaS platform that provided LLM evaluation, prompt management, and AI observability tooling to AI product teams at companies such as Gusto, Filevine, Dixa, FMG, Duolingo, and Vanta. Founded in 2020 and acquired by Anthropic in 2025, the platform was sunset in September 2025.
- Company typePrivate
- Founded2020
- HeadquartersSan Francisco, United States
- Headcount11–50
- GTM typeB2B
- OfferingSoftware
What Humanloop does
Humanloop was an enterprise SaaS platform that provided evaluation, prompt management, and observability tooling for teams building applications powered by large language models. The platform was purpose-built for the non-deterministic nature of generative AI outputs, offering a unified suite comprising four integrated modules: LLM Evaluations (supporting code-based, AI-powered, and human review workflows, plus LLM-as-a-Judge), Prompt Management with version control and multi-environment deployment (staging, production, QA, dev), AI Observability for production monitoring with alerting, guardrails, user feedback capture, and RAG system replay, and enterprise Compliance and Security features (SOC-2 Type 2, HIPAA, GDPR, VPC deployment, SSO/SAML, RBAC). Native integrations included the Vercel AI SDK, and the platform exposed a Python API for programmatic access to evaluations, datasets, logging, and prompt management.
The company served AI product teams at enterprise and growth-stage companies, including Gusto (HR/Payroll), Filevine (Legal Technology), Dixa (Customer Service), FMG (Financial Services), Athena (AI Assistants), Duolingo (Language Learning), Vanta (Security Compliance), and Twain (Sales). Customer outcomes cited included tripling AI deflection rates at Gusto, nearly doubling ARR at Filevine through six AI product launches in a year, shipping AI features 3x faster at Dixa, and achieving a 93% cost reduction at FMG by fine-tuning smaller models to match GPT-4 performance. Humanloop operated on a subscription-based enterprise SaaS model with quote-based annual contracts, supported by a hybrid go-to-market combining field sales with a "Book a demo" CTA for enterprise accounts alongside self-serve platform capabilities for smaller teams. The company was based in London, UK with US incorporation as Humanloop, Inc.
Founded in 2020 by Raza Habib (CEO), Jordan Burgess (CPO), and Peter Hayes (CTO)—all with PhD-level machine learning credentials from UCL and Cambridge and prior experience at Google AI, Amazon, and Monolith AI—Humanloop raised $2.6 million in seed funding from Index Ventures (lead), Y Combinator, LocalGlobe, and Albion VC in July 2022. The company was acquired by Anthropic in 2025, with the Humanloop platform sunset on September 8, 2025 and the founding team joining Anthropic's UK operations.
Humanloop firmographics
Firmographics- Name
- Humanloop
- Legal name
- Humanloop, Inc.
- Website
- https://humanloop.com
- Company type
- Private
- Founded year
- 2020
- Operating status
- Acquired
- Headcount range
- 11–50 employees
- Short description
- Humanloop was an enterprise SaaS platform that provided LLM evaluation, prompt management, and AI observability tooling to AI product teams at companies such as Gusto, Filevine, Dixa, FMG, Duolingo, and Vanta. Founded in 2020 and acquired by Anthropic in 2025, the platform was sunset in September 2025.
- Ownership category
- akta.pro rank
Humanloop industry classification
Industry- Product category
- LLM Evaluation Platform
- NAICS
- Custom Computer Programming Services (541511), Computer Systems Design and Related Services (54151)
- SIC
- Services-Prepackaged Software (7372), Services-Computer Programming Services (7371), Services-Computer Programming, Data Processing, Etc. (7370)
- akta.pro primary industry
- LLMOps & Generative AI Platforms (Prompt/Agent Orchestration, RAG) (HDAEANAD)
- akta.pro secondary industry
- MLOps/LLMOps & Model Lifecycle Management Services (BPAEAHAH)
Keywords
Where Humanloop is headquartered
LocationHeadquarters
- HQ city
- San Francisco
- HQ country
- United States
- HQ region
- North America
Offices1 record
Markets served
Humanloop business model
Business model- GTM type
- B2B
- Offering type
- Software
- Cost components
- Personnel, Technology or R&D, Marketing or Sales, Infrastructure, Operations
Revenue model
- SaaS Subscription: Enterprise SaaS platform with subscription-based pricing for LLM evaluation, prompt management, and AI observability tools. Platform sunset in September 2025 following acquisition by Anthropic.
Pricing tiers
| Model | Billing | Price |
|---|---|---|
| Subscription | Annual | Enterprise SaaS platform |
Go-to-market motion2 records
Distribution channels2 records
Marketing channels6 records
Humanloop product offering
Product offeringCore offering
Humanloop provides an enterprise SaaS platform for evaluating, managing, and monitoring large language model (LLM) applications. The platform combines LLM evaluations, collaborative prompt management with version control, production AI observability with guardrails and tracing, and enterprise compliance/security features to help product teams systematically build and ship AI-powered products.
Product overview
Humanloop was an LLM evaluation platform for enterprises, structured as a unified platform with four integrated modules: LLM Evaluations for assessing model quality, Prompt Management for version-controlled prompt development and deployment, AI Observability for production monitoring and logging, and Compliance & Security for enterprise security requirements. The platform enabled collaboration between engineers, product managers, and domain experts in AI development. Following its acquisition by Anthropic in 2025, the Humanloop platform was sunset with operations ceasing on September 8, 2025.
Differentiator
Problem solved
Functional benefit
Products and services
- LLM Evaluations Evaluation platform enabling enterprise AI teams to systematically assess large language model application quality through automated AI evaluations, code-based evaluators, human review, and CI/CD integration to prevent regressions before deployment.
- Prompt Management Platform for developing, versioning, and managing prompts and agents in code or UI. Includes interactive editors, version control for prompts/datasets/evaluators, multi-environment deployment, and integration with development workflows.
- AI Observability Production monitoring and logging platform for AI applications in production, providing alerting, guardrails, online evaluations, tracing, user feedback capture, and RAG system debugging with replay for ongoing reliability.
- Compliance and Security Enterprise security and compliance package including VPC deployment, EU or US cloud hosting, SOC-2 Type 2 certification, HIPAA compliance via BAA, GDPR compliance, RBAC, custom SSO/SAML authentication, third-party certified penetration testing, and TLS/SSL encryption with AES-256 at rest.
Quantifiable outcome
- Gusto tripled AI deflection rates and saved millions in support costs
- +4 more outcomes
Companies that use Humanloop
Customer profileNamed customers8 records
Segments3 records
Ideal customer profiles2 records
Humanloop technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- Yes
- API docs
- API detail
Core technology
AI maturity
App detail
Integration1 record
AI capability4 records
Feature4 records
Humanloop partnerships and signals
Strategic signalPartnerships
One partnership is on record.
- AnthropicflagshipHumanloop team joined Anthropic in 2025. The acquisition brought Humanloop's enterprise AI tooling expertise to Anthropic for building AI future that benefits everyone. Anthropic recruited leaders from HumanLoop's team for their UK operations.
Scale indicators3 records
Recent moves5 records
Expansion highlights4 records
Humanloop competitors and assessment
Company assessmentDirect peers
- Helicone: Open-source LLM observability platform offering logging, monitoring, and evaluation for AI applications. Direct competitor with comparable developer-focused positioning and pricing.
- Langfuse: Open-source LLM observability and evaluation platform offering tracing, prompt management, and evaluation tooling. Direct competitor to Humanloop's core offering; explicitly listed as a migration target for Humanloop customers after the platform sunset.
- Arize AI: Provides LLM observability and evaluation tooling including tracing, online evaluations, and production monitoring. Highly comparable feature set to Humanloop's observability product.
- LangSmith (LangChain): LangChain's commercial platform for debugging, testing, evaluating, and monitoring LLM applications. Directly comparable to Humanloop's evaluation and observability stack, integrated with the popular LangChain framework.
- Braintrust: LLM evaluation and observability platform for enterprises building AI applications. Competes head-to-head with Humanloop on prompt management, evals, and production monitoring; also listed as a Humanloop migration target.
Broad incumbents
- Datadog: Large-scale observability platform with emerging LLM monitoring and AI workload observability features. Notably, Datadog CEO Olivier Pomel was an advisor to Humanloop, indicating category overlap.
- Weights & Biases: Established MLOps platform with experiment tracking, model registry, and increasingly LLM evaluation capabilities. Broader than Humanloop but overlaps significantly in ML lifecycle tooling for AI teams.
Emerging players
- Confident AI (DeepEval): Open-source LLM evaluation framework (DeepEval) with commercial offering focused on benchmarking and testing LLM applications. Comparable in evaluation focus but smaller and more specialized than Humanloop's full platform.
Market position
Strengths5 records
Weaknesses5 records
Competitive moat4 records
Key risks7 records
Key highlights7 records
Customer concentration
Humanloop social profiles
Digital presenceHumanloop compliance and trust
Trust signalCompliance4 records
Humanloop financial estimates
Financial estimateRevenue estimate
Valuation estimate
Humanloop leadership team
Management profileNumber of profiles
Profiles5 records
Humanloop funding detail
Funding detailFunding overview
Funding rounds3 records
Investors5 records
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
Humanloop M&A and investment
M&A and investmentM&A
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about Humanloop
What does Humanloop do?
Humanloop provides an enterprise SaaS platform for evaluating, managing, and monitoring large language model (LLM) applications. The platform combines LLM evaluations, collaborative prompt management with version control, production AI observability with guardrails and tracing, and enterprise compliance/security features to help product teams systematically build and ship AI-powered products.
Is Humanloop a public or private company?
Humanloop is a private company. It is classified as corporate owned and is currently acquired.
When was Humanloop founded?
Humanloop was founded in 2020. It employs 11 to 50 people.
Where is Humanloop based?
Humanloop is headquartered in San Francisco, United States, in the North America region.
How does Humanloop make money?
One revenue line is on record: saaS Subscription.
Who are Humanloop's main competitors?
Direct peers on record are Helicone, Langfuse, Arize AI, LangSmith (LangChain) and Braintrust. Broad incumbents are Datadog and Weights & Biases. Confident AI (DeepEval) is listed as an emerging player.
Does Humanloop have an API?
Yes. Humanloop provides a public API enabling developers to programmatically interact with the platform. The API supports key operations including: running evaluations via API, creating datasets via API, logging through API, managing prompts, tools, datasets, evaluators, and accessing evaluation results, runs, and statistics. Users access the API using API keys from the API Keys page (app.humanloop.com/account/api-keys). The platform also supports CI/CD integration for automated evaluations. Developer documentation is at humanloop.com/docs.
What industry is Humanloop in?
Humanloop's product category is LLM Evaluation Platform. Its primary akta.pro industry code is HDAEANAD, LLMOps & Generative AI Platforms (Prompt/Agent Orchestration, RAG), with a secondary code of BPAEAHAH, MLOps/LLMOps & Model Lifecycle Management Services. Its NAICS code is 541511 and its SIC code is 7372.