Data Product
Training data drawn from real expert work
Expert datasets, trajectories, and task environments for post-training, agent training, and targeted evaluation — built with the specialists who do the work.
Product overview
WebDev
Expert-evaluated front-end builds paired with briefs, development trajectories, and rubric-based scores for aesthetics, motion, and interaction.
Vision2Web
Runnable front-end and full-stack implementations paired with visual references, interaction specifications, PRDs, and programmatic evaluation.
Web3D
A multi-category corpus of interactive 3D object assets — programmatic models, functional assemblies, deterministic behaviors, and expert-defined evaluation.
Finance Long-Horizon Agent Tasks
A structured dataset of expert-curated, long-horizon finance tasks designed to train and evaluate AI agents on real-world financial research, modeling, analysis, and deliverable production.
Build the dataset your model is missing.
Every product here is built with domain experts and delivered to your training or evaluation spec. Tell us what you need.
Discuss your project