Research

We build the environments and benchmarks that teach AI agents to do real financial work—the documents, models, and multi-step judgment calls behind the most economically valuable tasks—and we publish what we learn about where frontier models succeed and where they break.

Long-Horizon Finance Evaluation

Westworld Finance Diligence Bench Evaluating AI Agents on a Complete Due Diligence Process

DealTrace Bench A staged review of a real private-equity deal in five graded stages—extract, reconcile, forecast, market, recommend.

Agent Benchmarks

Web Bench A benchmark of ~2,454 realistic tasks across 452 real-world websites for evaluating browser agents.

BrowserBench How bot-detection systems distort browser-agent evaluation, with stealth vs. non-stealth performance across real infrastructure.

RL Environments

Simulated-to-Real Transfer for Web Agents: A small scale study Does training a web agent on a simulated Best Buy transfer to the real one? A simulated-to-real small study with Westworld's Goodbuy clone.

Westworld v1 AI sandbox environments for training and evaluating computer-use agents on realistic simulated websites.