A collection of thoughts and announcements
Westworld Finance Diligence Bench
Evaluating AI Agents on a Complete company acquisition due-diligence process
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
DealTrace
Tracing where frontier models break on a long private-equity deal review
Westworld v1
Benchmarking and Training Web Agents in Highly Realistic Simulators via Verifiable Rewards
BrowserBench
The First Benchmark for Browser Infrastructure Stealth
Web Bench: The Current State of Browser Agents
Benchmarking browser agents on ~2.5k tasks across 452 websites