Louis Manhès

Founder and machine-learning engineer. I build applied AI products, and I research symbolic autotelic agents: machines that set their own goals and learn the structure of their world as laws they can explain.

Get in touch

Selected work

Products I have built or am building, and the research alongside them.

Genario

Active

AI software and services for screenwriting and audiovisual production

ScreenwritingAudiovisualScript analysis

2019

AntiAgent

Active

An AI notebook where pages teach you back

LearningSpaced repetitionFSRS

2026

Eliane

Active

Prescription coordination for French home-care nurses

HealthcareNursingSaaSCompliance

2026

Cubist

Research

Building a symbolic autotelic agent: an AI that sets its own goals and learns its world as readable laws

Autotelic agentsSymbolic descentWorld-modelsOpen-endednessARC-AGI-3Interpretability

2026

All projects

Writing

Notes on the systems I build and the research behind them.

Jul 3, 2026· 12 min read

Laws from single experiences: an online symbolic world-model for ARC-AGI-3

Cubist's world-model learns each ARC-AGI-3 game's mechanics as a small theory of symbolic laws, online, from single experiences, with no pretraining and no gradients. This post presents the representation, the learning algorithm, and an evaluation across all 25 public games.

Jul 1, 2026· 10 min read

Symbolic descent: gradient descent, applied to rules instead of weights

Symbolic descent keeps the shape of gradient descent but changes the object it optimizes, from a weight vector to a readable theory of laws. This post explains the parallel, the machinery that makes it work, and what it buys for continual learning, reasoning, and interpretability.

All writing