Researcher · Engineer

Jeremy Berman

AnthropicReinforcement Learning

Berman built evolutionary systems that achieved leading scores on ARC-AGI, a benchmark of adapting to unfamiliar reasoning puzzles. His approach generates and refines candidate strategies while solving a problem, exploring how models can discover answers rather than recall patterns.

Career & education

Current role

AnthropicSep 2026-present

Reinforcement Learning

Previously

Humans&Nov 2025-Aug 2026

Member of Technical Staff, Founding Team

Reflection AIAug 2025-Nov 2025

Member of Technical Staff

NdeaJan 2025-Aug 2025

Member of Technical Staff

Education

University of Pennsylvania

Work and education history is primarily focused on AI-relevant roles and may not be comprehensive.

Updates 1

MoveJeremy Berman joins Anthropic

The ARC-AGI researcher joined Anthropic from Humans& to work on reinforcement learning for Claude.

September 8, 2026

About this profile

Based on public sources. If accuracy is important, verify details using the links or references.

For corrections, updates, or requests to remove information, contact us.

Reviewed September 20, 2026 · Checked for accuracy and new information.