Teaching Machines to Think Like Us

Most AI systems excel at narrow tasks — classifying images, generating text, predicting outcomes from structured data. But they fail spectacularly when faced with the messy, ambiguous, context-dependent decisions that humans handle effortlessly every day. Simile is attacking this gap head-on.

Their $100 million Series A, led by Index Ventures with participation from Hanabi Capital and Bain Capital Ventures, reflects growing enterprise demand for AI that can handle real-world complexity — not just textbook problems. The technology builds "decision twins" that capture the reasoning patterns of expert humans, then applies those patterns at scale.

Where It's Being Used

Early deployments span financial services (replicating how top traders assess risk), healthcare (capturing how experienced clinicians prioritize patient care), and supply chain management (mimicking how veteran logistics managers handle disruptions). The common thread: domains where getting it right matters more than getting it fast, and where human expertise has traditionally been the only reliable guide.