Company index · data as of 2026-10-15

AI data companies

Public research on 39 AI data organizations: business model, revenue, funding, valuation and paid work routes, each with dated sources.

Open the interactive index

• A.Team · Managed AI & data services
  Embeds senior engineering teams to build and operate enterprise agentic AI systems
• Abundant · Frontier data & RL environments
  Builds simulation environments and datasets that train safer, more capable AI agents.
• AfterQuery · Expert data & evaluation
  AfterQuery builds expert-generated datasets and environments for frontier-model training. Its enterprise team also develops custom evaluations and deploys systems on a company’s own data.
• AIXI Labs · Nonprofit research
  Researches mathematical foundations of powerful agents and practical mitigations for AI risk
• Alinia AI · AI evaluation & safety
  Builds policy judges that monitor, test, and control regulated enterprise AI agents
• Appen · Managed AI & data services
  Delivers human-validated datasets for frontier models, agents, speech, and embodied AI systems.
• APTO · Managed AI & data services
  Creates multimodal training data through expert annotation services and proprietary harBest software
• BenchFlow · Frontier data & RL environments
  Builds stateful environments, benchmarks, and runtime infrastructure for training capable AI agents.
• Bespoke Labs · Frontier data & RL environments
  Builds realistic training environments and optimization infrastructure for reliable autonomous AI agents.
• Datacurve · Frontier data & RL environments
  Recruits expert developers to create challenging training datasets and evaluate coding agents.
• Elorian · Compute & model research
  Develops foundation models that reason visually about space, physics, and creative problems.
• G2i · Managed AI & data services
  Supplies senior engineers building RL environments, evaluations, and high quality training data
• Giskard · AI evaluation & safety
  Tests enterprise AI agents for security failures, hallucinations, and other reliability risks
• HackerRank Bench · AI software & analytics
  Sells developer assessments alongside coding datasets, private evaluations, and reinforcement learning environments
• Handshake AI · Expert data & evaluation
  Mobilizes professional experts to create training data, evaluations, and environments for AI.
• iMerit / EXL · Managed AI & data services
  Combines specialist annotators and software for multimodal training data and model evaluation.
• Invisible Technologies · Managed AI & data services
  Combines expert data, evaluations, and configurable AI platforms to automate enterprise operations.
• Labelbox · Frontier data & RL environments
  Builds enterprise agents, reinforcement-learning environments, robotics data and expert training signals.
• Laude Institute / Terminal-Bench · Nonprofit research
  Funds open computing research and develops benchmarks evaluating realistic terminal agent capabilities
• Mercor · Expert data & evaluation
  Turns professional expertise into training data, evaluations, enterprise agents and licensed datasets.
• micro1 · Expert data & evaluation
  Builds expert human datasets, realistic training environments, and contextual evaluations for agents.
• Ooak Data · Frontier data & RL environments
  Turns authentic company workflows into anonymized environments for training enterprise AI agents.
• Pareto AI · Expert data & evaluation
  Turns specialist human judgment into training data and evaluations for frontier models.
• Polymath · Frontier data & RL environments
  Builds realistic software worlds to train autonomous agents across long engineering workflows.
• Prime Intellect · Compute & model research
  Provides compute, training, inference, and environments for teams building their own models
• Prolific · Expert data & evaluation
  Connects researchers and AI teams with verified participants supplying paid human feedback
• Proximal · Frontier data & RL environments
  Builds research-driven data systems and demanding benchmarks for autonomous technical AI agents.
• Quesma · AI software & analytics
  Analyzes coding-agent sessions to explain costs, expose failures, and improve engineering outcomes.
• Refresh · Frontier data & RL environments
  Creates simulated software worlds and Blender tasks with verifiable rewards for agents.
• RWS / TrainAI · Managed AI & data services
  Delivers multilingual training data, annotation, validation, and expert feedback for enterprise AI
• Sama · Managed AI & data services
  Provides human-verified annotation, validation, and evaluation for scalable production-ready multimodal AI systems.
• Scale AI · Managed AI & data services
  Supplies expert training data, evaluations, and deployed AI for enterprises and governments.
• Snorkel AI · Frontier data & RL environments
  Develops specialized datasets, evaluation systems, and executable environments that advance frontier AI.
• Surge AI · Expert data & evaluation
  Creates expert human data, feedback, evaluations, and environments for frontier AI training.
• TaskUs · Managed AI & data services
  Provides outsourced customer operations, trust and safety, and specialized AI data services.
• Terac · Expert data & evaluation
  Matches verified domain experts with paid research, evaluation, and AI training projects
• Toloka · Managed AI & data services
  Provides expert human data, evaluations, and environments for training reliable AI systems
• Trajectory Labs, PBC · AI evaluation & safety
  Builds safety evaluations, adversarial tests, and training environments for frontier AI systems.
• Turing · Expert data & evaluation
  Supplies expert datasets and environments while deploying enterprise AI across complex workflows.