Agent evaluation.
RL environments.

We provide evaluation data and RL environments to improve personal agents across apps. Tasks are designed using real user insights.

  1. 01/03Users

    What does this user need?

    Tasks and constraints built from real user insights and synthetic personas.

  2. 02/03Evaluate

    Watch the task. Find the failure.

    Check task completion, preferences, and permissions.

    Domains

    Dining · Travel · Shopping · Money · Health · Home · Work · and more

    MemoryPersonalizationProactivitySafety & securityRetrievalExecutionand more
  3. 03/03Train

    Train in connected environments.

    Digital worlds, user simulators, and rewards for intermediate steps and final outcomes.

The data we deliver.

01◎

Evaluation Tasks

Tasks with user goals, app context, and success criteria to test your agent.

Explore ↗
02≋

SFT data

User-agent interaction trajectories for supervised fine-tuning.

Explore ↗
03⌘

RL environments

Apps, tools, user context, and task rewards for training your agent.

Explore ↗

MatrAIx: A data and evaluation lab leading the personal agent era.

Contact Us ↗