Samuel Berthelot

Pretrains a small Llama-style language model, written from scratch in PyTorch, on the TinyStories dataset, then uses it to fit a scaling law and to try the Muon optimizer and a mixture of experts.

Python, PyTorch, Hugging Face Tokenizers, NumPy, Matplotlib

Treats forecasting as language modeling, like the Chronos paper: a small transformer trained on real and synthetic series predicts each next value as a token, and is tested on two real datasets it has never seen, ETTh1 and ETTm1.

Python, PyTorch, NumPy, Matplotlib

Classifies breast ultrasound images as normal, benign or malignant with a ResNet-34 and segments the lesion with a U-Net, then tests both models on a dataset from another hospital that they never see in training.

Python, PyTorch, torchvision, scikit-learn, SciPy, Jupyter

Passes an English sentence through a chain of NLLB-200 translations, translates each step back to English and measures how far the meaning drifts, like a game of telephone.

Python, PyTorch, Hugging Face Transformers, sacreBLEU, NLTK, scikit-learn

Races a simulated car with model predictive control, then corrects the controller's imperfect car model with a neural network trained on its prediction errors, all on an autodiff engine written from scratch in NumPy.

Python, NumPy

Evolves neural networks that learn to play Snake with a genetic algorithm. It started as my TIPE, the research project presented at the end of French classes préparatoires; the 2026 version adds tests and measured results.

Python, NumPy, pygame, pytest

Asks whether monthly absence hours at a courier company in Brazil can be forecast (740 absences, July 2007 to July 2010), then explains and classifies absence durations instead.

Python, pandas, statsmodels, scikit-learn, Jupyter, LaTeX

Simulates liquids on a grid with a cellular automaton in C and SDL3: draw walls, pour 20 liquids with realistic densities, and watch them flow, mix and build up pressure.

C, SDL3

A chess game in Rust with its own engine and no chess library: negamax search with alpha-beta pruning behind a macroquad interface.

Rust, macroquad

Runs Ruby code from a browser editor on the local machine, starting a new Docker or Podman container with the chosen Ruby version for each run and deleting it afterwards.

JavaScript, React, Monaco Editor, Node.js, Express, Docker