Skip to content
View JeremyMaille's full-sized avatar

Block or report JeremyMaille

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.

Content in all repositories owned by your account will be closed.
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
JeremyMaille/README.md

Jérémy Maille

Machine Learning Engineer in training | AI & Data Specialization

Final-year engineering student at CESI École d'Ingénieurs

LinkedIn Email


🚀 What I build

I design and ship end-to-end ML systems: data pipelines, model training, evaluation, and API deployment. I care about measuring what actually works: my projects report baselines, ablations, and the techniques that did not help. My focus areas are deep learning for computer vision & time series, NLP/RAG systems, and MLOps.

📌 Featured Projects

Project What it does Stack Results
📋 RAG on FAA Regulations Evaluated RAG over aviation regulatory documents (ADs, 14 CFR): hybrid retrieval, query decomposition, reranking, agentic follow-up Qdrant, BM25, sentence-transformers, Mistral 70% → 95% recall@5 across 6 ablation stages · full failure analysis
🛰️ Urban Change Detection Siamese U-Net (ResNet34 encoder) detecting building changes between two satellite images, with a live demo PyTorch, TorchGeo, Gradio F1 0.536 on LEVIR-CD+ (+464% vs non-ML baseline) · Live demo 🤗
✈️ Flight Trajectory Prediction Transformer-based model predicting aircraft trajectories, served via REST API PyTorch, Transformers, FastAPI, Docker Deployed inference API
⚙️ Predictive Maintenance / RUL CNN-LSTM hybrid predicting Remaining Useful Life of turbofan engines TensorFlow/Keras, NASA C-MAPSS RMSE 37.74 cycles + scientific poster
🧠 Mouse Behavior ML Baselines Full supervised-learning pipeline for behavioral analysis, research @ CERVO Brain Research Centre Python, scikit-learn, CV Academic research project
🎓 NovaCampus RAG Assistant Retrieval-Augmented Generation assistant inside a microservices academic ERP Qdrant, NestJS, LLM APIs, Docker 5-layer microservices architecture
🕹️ RL Pacman Agent Reinforcement learning agent trained to play Pacman Python, RL Learning project

🛠️ Tech Stack

ML / Deep Learning PyTorch scikit-learn Hugging Face MLflow

Data & Backend Python Pandas FastAPI PostgreSQL MongoDB

MLOps & Infra Docker Git Linux

🎓 Background

  • Master's degree in Computer Science, AI & Data specialization, CESI École d'Ingénieurs (2026)
  • Research experience in Computer Vision, CERVO Brain Research Centre, Université Laval, Québec 🇨🇦
  • Previous internships in software engineering and cybersecurity

🔍 Currently seeking a 6-month end-of-studies internship starting January 2027 in ML Engineering / Data Science. Open to relocation in France and internationally.

Pinned Loading

  1. rag-aero rag-aero Public

    Evaluated RAG on FAA regulatory documents: ablation study across 6 retrieval stages, with measured regressions and failure analysis

    Python

  2. urban-change-detection urban-change-detection Public

    Siamese U-Net (ResNet34) for building change detection on satellite imagery · F1 0.536 on LEVIR-CD+ · Live demo on HF Spaces

    Jupyter Notebook

  3. flight-trajectory-prediction flight-trajectory-prediction Public

    Transformer-based aircraft trajectory prediction from live ADS-B data · end-to-end pipeline with FastAPI inference API + Docker

    Jupyter Notebook

  4. mathilde-ly/mouse-behavior-ml-baselines mathilde-ly/mouse-behavior-ml-baselines Public

    Research Project | CERVO Brain Research Centre This repository contains the full pipeline for studying mouse behavior in controlled environments using supervised learning. Developed under the super…

    Python

  5. PROJETWEB-CESI/Project-Web-Group1 PROJETWEB-CESI/Project-Web-Group1 Public

    JavaScript

  6. Reinforcement-Learning-Pacman Reinforcement-Learning-Pacman Public

    Python