Skip to content
View Mathieu-Le-Gouill's full-sized avatar
3 followers · 2 following

Block or report Mathieu-Le-Gouill

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Mathieu-Le-Gouill/README.md

Hi, I'm Mathieu 👋

Typing SVG

LinkedIn Portfolio Email views

🧠 About me

AI/ML engineer with a double degree in Artificial Intelligence (Engineering degree at ENIB + Master's at UQAC, Canada). I design robust, reproducible, actually-deployed AI systems - from SIMD-optimized deep learning libraries in modern C++ to containerized RAG / LLM pipelines in the cloud.

📍 Brest, France  ·  🗣️ French (native) · English (C1) · Spanish · Mandarin (notions)

🚀 Featured Projects

⚡ Deep Learning Library in C++

A deep learning library built from scratch in modern C++17: compile-time tensor pipelines (variadic templates), core ops (dense, activation, loss, optimizer) and low-level SIMD vectorization to max out single-core CPU throughput.

C++17 · SIMD · CMake · Templates

🌐 Transformer EN→FR - from scratch

A complete Transformer architecture implemented from scratch in PyTorch for sequence-to-sequence translation, trained on OPUS Books with pretrained BERT / CamemBERT tokenizers. Train/inference pipeline driven by Makefile.

Python · PyTorch · CamemBERT · NLP

🔎 LangChain RAG - REST API

A production-ready RAG system querying a vector store to return context-enriched answers, exposed through a modern REST API and fully containerized with Docker for reproducible deployment.

Python · LangChain · FastAPI · Docker

🎨 VQ-VAE & Generative Models Suite

From-scratch implementations of VQ-VAE, β-VAE and standard VAE, trained on MNIST for image generation & reconstruction - exploring discrete latent spaces and codebook learning.

Python · PyTorch · Generative AI

all repos

🛠️ Tech Stack

Languages

Python C++ Java Rust HTML/CSS

Deep Learning & Generative AI

PyTorch TensorFlow scikit-learn LangChain RAG

MLOps & Cloud

Docker Kubernetes FastAPI GCP AWS Azure ML

HPC / Systems & Tools

SIMD CUDA Git CMake Linux Qt

Building AI systems from low-level C++ kernels up to cloud-deployed LLM pipelines.

Popular repositories Loading

  1. Deep_Library_Cpp Deep_Library_Cpp Public

    Custom C++ deep learning library optimized with SIMD for enhanced CPU performance.

    C++ 1

  2. Image_Denoising_With_Autoencoders Image_Denoising_With_Autoencoders Public

    Jupyter Notebook

  3. Transformer-EN-FR Transformer-EN-FR Public

    Custom Transformer implementation for english to french translation using pretrained tokenizer and the dataset OPUS

    Python

  4. LangChain_RAG LangChain_RAG Public

    Retrieval Augmented Generation pipeline with LangChain, FastAPI and Uvicorn

    Python

  5. Vector_Quantized_Variational_Autoencoder Vector_Quantized_Variational_Autoencoder Public

    Custom implementation of VQ-VAE and similars models like Beta_VAE, VAE and simple AE. Models training and testing on MNIST dataset.

    Python

  6. Mini-Git Mini-Git Public

    A minimal Git-like version control system written in Python.

    Python