What micro1 does

micro1 is a Palo Alto-based AI data research company that builds infrastructure for training and evaluating frontier AI systems. It operates a proprietary platform combining a vetted network of more than two million experts across 71 countries with purpose-built tooling, supplying expert annotation, egocentric robotics video, RL environments, and contextual evaluation services to AI labs and Fortune 100/500 enterprises. Its named product suite includes Realm (RL environments for agentic training), Cortex (in-production AI agent evaluation), Robotics Data (high-fidelity data for humanoid robots), and the AI Recruiter Agent (originally branded Zara), supported by an Interview Prep tool, a Human Data Platform, and a research division publishing benchmarks such as LongExtractBench and RedlineBench. Named enterprise customers include Microsoft, OpenAI, Google, Box, Physical Intelligence, and Boston Dynamics.

The company was founded in 2022 by Ali Ansari initially as an AI recruitment platform under the Zara brand, then pivoted in 2025 to AI training data and evaluation infrastructure, rebranding to micro1. Following a $3.3M seed round in mid-2024 (at ~$80M valuation) and a $35M Series A led by 01 Advisors in September 2025 (at a $500M valuation), the company reached $100 million in ARR by December 2025, with investors subsequently offering a $2.5 billion valuation. CEO Ali Ansari reportedly holds approximately 42% of the company, and former Twitter COO Adam Bain joined the board as part of the Series A.

Micro1's business model blends managed services (paid human annotation and data labeling), usage-based expert fees (up to $500/hour for top domain experts, ~$200/hour for general AI trainers, $15/hour for robotics video contractors), and subscription/recurring platform revenue. Go-to-market is hybrid: direct enterprise sales into AI labs and Fortune 100 buyers, complemented by a self-service portal for individual experts and interview-prep users, with supporting demand generation through content, research publications, and community events. Reported operational outcomes include 100% retention on competitive-programmer engagements, 250 annotators onboarded within 30 days for a major AI data provider, and 50% faster data generation than client expectations on AI trainer projects.

Name

micro1

Legal name

micro1 Inc.

Website

https://micro1.ai

Company type

Private

Founded year

2022

Operating status

Operating

Headcount range

101–250 employees

Short description

Micro1 builds AI training-data and evaluation infrastructure for frontier AI labs and Fortune 100 enterprises, combining a vetted network of 2M+ experts with RL environments, agent evaluation tooling, and robotics data collection.

Ownership category

akta.pro rank

Product category

AI Training Data & Evaluation Platform

NAICS

Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (518), Computer Systems Design and Related Services (54151)

SIC

Services-Computer Integrated Systems Design (7373)

akta.pro primary industry

Annotation Tooling & Workflow Management (HDAAALAC)

akta.pro secondary industries

AI Observability, Monitoring & Evaluation Platforms (Drift, Quality, Safety) (HDAEANAF), Data, Analytics & AI Training (BI, Data Engineering, ML/GenAI) (BPAEANAH), Audit, Explainability & Accountability Tooling (traceability, reporting) (HDAAAKAL)

Keywords

• AI training data
• Human annotation services
• Reinforcement learning environments
• AI agent evaluation
• Robotics data collection

Headquarters

HQ city

Palo Alto

HQ country

United States

HQ region

North America

Offices2 records

Markets served

GTM type

B2B

Offering type

Software

Cost components

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

Revenue model

1. AI Training Data Services: Provides training data for AI models by hiring domain experts to annotate and grade AI outputs for major AI labs and tech companies. Revenue from data labeling and management services.
2. Expert Network Platform: Network of over 2 million experts who help refine AI models. Professionals like medical and finance experts earn up to $500 per hour through the platform.
3. Robotics Data Collection: Collects video data from contractors performing household tasks for training humanoid robots. Data sold to robotics companies.

Pricing tiers

Go-to-market motion2 records

Distribution channels3 records

Marketing channels5 records

Core offering

micro1 operates a data research company that builds infrastructure for training and evaluating AI models: it recruits, vets, and manages a global network of 2M+ domain experts who produce supervised, RL, and evaluation data for frontier AI labs and Fortune 100 enterprises. Customers consume its offerings through three core product lines — Realm (RL environments for agentic training), Cortex (production evaluation for AI agents), and Robotics Data (egocentric video for humanoid robots) — delivered alongside managed expert annotation services and an AI-driven recruiting/evaluation engine.

Product overview

micro1 is a data research company providing infrastructure for advancing AI intelligence through expert human data, real-world training environments, and contextual evaluations. The platform operates as a unified system comprising multiple integrated products: Realm (RL environments for agentic AI training), Cortex (AI agent evaluation platform), Robotics Data (human movement data for humanoid robots), and the AI Recruiter Agent (Zara, for hiring vetted talent). Additional offerings include an Interview Prep tool with 150+ practice interview types, a Human Data Platform for managing expert annotation projects, and a network of 2M+ vetted experts spanning domains like law, medicine, and competitive programming. The company also publishes proprietary benchmarks including LongExtractBench and RedlineBench for evaluating AI systems. The portfolio positions micro1 as a comprehensive human data and AI evaluation infrastructure provider for frontier AI labs and Fortune 100 companies.

Differentiator

Problem solved

Functional benefit

Brands

• Realm: RL environments that mirror real-world scenarios to generate world-class human data for agentic actions and elevate model reasoning.
• Cortex
• Robotics

Products and services

• Realm Frontier evaluations and reinforcement-learning environments for AI labs that mirror real-world scenarios to generate expert human data for agentic actions and elevate model reasoning. Sold to AI labs building agentic and frontier models.
• Cortex Contextual evaluation platform that provides a visibility and improvement layer for enterprise AI agents in production, enabling performance monitoring, benchmarking, and continuous improvement. Sold to enterprise customers deploying AI agents.
• Robotics Data High-fidelity, expert-demonstrated egocentric video data service for training humanoid and embodied AI systems; sourced from a 71-country contractor network (4,000+ contractors contributing 160,000+ hours/month of household-task footage). Sold to frontier robotics labs such as Physical Intelligence and Boston Dynamics.
• AI Recruiter Agent (Zara) AI-powered recruitment engine that screens applicants, conducts interviews including coding exercises, and generates post-interview reports with dynamic, real-time skills evaluation at scale. Sold to enterprise customers and used internally to staff micro1's own expert network.
• Interview Prep Self-serve AI-powered practice interview platform offering questions and feedback for over 150 job positions across software engineering, healthcare, finance, legal, and other industries. Used by individual job seekers.
• Human Data Platform Proprietary managed-services platform for end-to-end human data generation projects, covering sourcing, vetting, onboarding, and quality assurance of subject-matter experts for AI training and evaluation work.
• Expert Network Network of over 2 million vetted experts including PhDs, MBAs, lawyers, doctors, and competitive programmers (IOI-level) available to AI labs and enterprise customers for AI training, RLHF, and evaluation projects.

Quantifiable outcome

• 87% reduction in financial costs compared to traditional hiring methods (AI-led interviews)
• +5 more outcomes

Named customers7 records

Segments4 records

Ideal customer profiles3 records

Technology focussed Yes

API detail

Has API

No

API docs

API detail

Core technology

AI maturity

App detail

AI capability11 records

Feature5 records

Partnerships

Two partnerships are on record, tiered minor.

• ReductominorStrategic or Co-development Partner · 1 July 2026Reducto's Deep Extract ranked first in LongExtractBench benchmark released by micro1 for document extraction evaluation
• CrosbyminorStrategic or Co-development Partner · 17 June 2026Partnership to release RedlineBench benchmark for evaluating AI systems on contract redlining and SaaS negotiation

Scale indicators10 records

Recent moves6 records

Expansion highlights6 records

Direct peers

• Labelbox: Labelbox provides a data-labeling platform plus managed annotation services for AI teams. It overlaps with micro1's Human Data Platform on tooling, vendor workflows, and AI lab customers.
• Surge AI: Surge AI (now part of Scale) is a human-data platform focused on expert annotation and RLHF for foundation models. It overlaps directly with micro1's expert annotation and evaluation offerings for frontier AI labs.
• Scale AI: Scale AI is the dominant data-labeling and RLHF provider for frontier AI labs. It is micro1's most direct competitor, and the explicit "Scale AI exodus" by Microsoft, OpenAI, and Google is the engine behind micro1's growth.
• CloudFactory: CloudFactory offers managed data annotation and AI data services via a distributed workforce. It targets AI/ML teams needing human-labeled data at scale, similar to micro1's expert and robotics workflows.
• Invisible Technologies: Invisible Technologies provides AI training data and digital operations workforces for enterprise and frontier-model customers. It competes head-to-head with micro1 in expert annotation, RLHF, and managed human-data operations.

Broad incumbents

• Telus International AI: Telus International AI (formerly Lionbridge AI) is a large, established data annotation and digital CX provider serving AI teams and enterprises. It competes with micro1 on volume data labeling but with a less expert-focused, lower-margin model.
• Appen: Appen is a publicly listed data annotation and dataset company with a global crowd. It serves many of the same AI labs and enterprises as micro1 but with a much broader, less expert-skewed model and slower product iteration.
• AWS SageMaker Ground Truth: AWS SageMaker Ground Truth is the hyperscaler data-labeling offering integrated with the broader AWS ML stack. It is an indirect competitor and potential partner given its commoditized positioning versus micro1's expert-led, RLHF-grade offering.

Emerging players

• Datasaur: Datasaur provides an NLP-focused data labeling and LLM fine-tuning platform. It targets similar AI/ML teams with labeling workflows and overlaps with micro1 on managed annotation services for text and document AI use cases.
• Snorkel AI: Snorkel AI focuses on programmatic data labeling and weak supervision for foundation models. It competes with micro1 on the AI-training-data layer, particularly where enterprises want less reliance on human annotators.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat5 records

Key risks6 records

Key highlights7 records

Customer concentration

• LinkedInlinkedin.com/company/micro1
• Xx.com/micro1_ai

Revenue estimate

Valuation estimate

Number of profiles

Profiles2 records

Funding overview

Funding rounds4 records

Investors12 records

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 micro1