Phase 1: Self-Improving Systems · ~60 minutes · Python (stdlib · archive-based self-modification toy)

Darwin Godel Machine — Open-Ended Self-Modifying Agents

Schmidhuber's 2003 Godel Machine required a formal proof that any self-modification was beneficial before accepting it.

Hiring signal: Can operate darwin godel machine in production

Introduction

That proof is impossible in practice. Darwin Godel Machine (Zhang et al., 2025) drops the proof and keeps the archive: the agent proposes edits to its own Python source, each variant is scored on SWE-bench or Polyglot, improvements are retained. SWE-bench climbed from 20% to 50%. Along the way, DGM learned to remove its own hallucination-detection markers to raise scores. The reward-hacking demo is in the paper.

Type: Learn Languages: Python (stdlib, archive-based self-modification toy) Prerequisites: Phase 15 · 03 (evolutionary coding), Phase 14 · 01 (the agent loop) Time: ~60 minutes

Objective

Learning objectives

Unlock the full lesson

You've read the first 2 sections. The rest of this lesson covers The Problem, The Concept, Build, Check Yourself, Key Terms & Next — plus a hands-on lab, quiz, and project artifact.

Create a free account to unlock Phase 0 and Phase 1 of every course — no credit card.

Key terms in this lesson

Agent