Killian Steunou

About Me

I'm a second-year master's student in Mathématiques, Vision, Apprentissage at ENS Paris-Saclay. I'm passionate about combining mathematics, statistics, and cutting-edge artificial intelligence. My academic focus revolves around deep learning, particularly in image processing.

Beyond theory, I'm a dedicated coder. I constantly explore new AI algorithms and techniques, pushing the boundaries of what's achievable. I am always trying to improve my technical skills and deepen my understanding for applying mathematical concepts in the real world.

In my projects, I blend robust statistics with innovative AI to uncover valuable insights. My aim is to contribute to AI by leveraging my strong math foundation, using algorithms to solve complex problems and drive transformative technologies.

This portfolio showcases my journey, highlighting projects that demonstrate my skills and passion. It reflects my academic pursuits and reveals my aspirations in the exciting field of AI and applied mathematics.

Professional Experiences

Incoming Intern - AI Research - Idemia

April 2025 - October 2025

Research on Multi Object Tracking.

Intern - AI Research - CLS

April 2024 - August 2024

In charge of evaluating the relevance of using Foundation Models for remote sensing applications. For this purpose, I developed a library to fine-tune Foundation Models on remote sensing datasets, used by coworkers at CLS. Key tasks include:

Literature Review: A comprehensive review of the latest research on Foundation Models and their applications in remote sensing. This involved analyzing existing studies, identifying gaps in the literature, and determining the potential benefits of using Foundation Models for remote sensing tasks.
Model Development: I developed a library to fine-tune Foundation Models on remote sensing datasets. This included implementing data preprocessing pipelines, training and evaluation scripts, and fine-tuning algorithms to adapt Foundation Models to remote sensing tasks.
Performance Evaluation: I evaluated the performance of fine-tuned Foundation Models on semantic segmentation, which is a key task in remote sensing, by comparing key metrics and the computation time to performance ratio with traditional models trained from random weights.

Intern - Machine Learning Engineer - JoliBrain

February 2023 - July 2023

Contribution to joliGEN: I implemented a range of advanced edge detection methods to significantly enhance the capabilities and usability of the open-source tool joliGEN. This included incorporating image generation controls (from the ControlNET paper) to provide users with greater flexibility and control over their image generation tasks.
I implemented state‐of‐the‐art machine learning models such as Owl‐ViT for zero‐shot object detection.
I took charge of developing the entire documentation website for the tool, using ReadTheDocs. This involved creating a user-friendly, comprehensive resource that enabled users to quickly understand and utilize the tool's features and functionalities, enhancing the overall user experience.

Intern - Software Developer (R Shiny) - French Ministry of Agriculture

May 2022 - August 2022

I created agreste : a comprehensive automation tool as an R package for streamlining the creation of statistical publications.
I took ownership of the project from initial requirements gathering to final tool delivery, using Agile methodologies for efficient project management.
I designed an intuitive user interface using R Shiny to facilitate data input and publication generation, enhancing user experience.

Academic Background

Master 2 Mathématiques, Vision, Apprentissage - ENS Paris-Saclay

September 2024 - March 2025

Computational Optimal Transport, Convex Optimization, Object Recognition and Computer Vision, Probabilistic Graphical Models, Deep Learning and Signal Processing, AI Safety and Explainability.
Generative Models for Image, 3D Modeling, Representation Learning for Computer Vision, Remote Sensing Data, Deep Learning for Image Synthesis.

Master 1 Applied Mathematics and Statistics - Toulouse School of Economics

September 2023 - April 2024

Econometrics, Probability and Statistics, Functional Analysis, Convex Optimization for Machine Learning, Computational Data Science (Python).
Foundations of Machine Learning, Mathematical Game Theory, Advanced Optimization, Markov Decision Processes, High Dimensional Data Analysis and Machine Learning, Time series, Stochastic Methods for Optimization and Sampling.

Gap Year - University of Copenhagen

September 2022 - January 2023

I took courses in Natural Language Processing, Blockchain Business Development, Energy Economics, and Tax Policy.

Double Bachelor: Applied Mathematics and Economics - Toulouse School of Economics

September 2019 - April 2022

Linear Algebra, Analysis, Probability and Statistics, Numerical Analysis, Optimization, Econometrics, Computer Science (Python, R), Microeconomics, Macroeconomics.

Some of my Recent Projects

Video Background Removal

A video background removal tool using Mobile SAM (Segment-Anything Model) to automatically segment and remove the background from videos.

Check it Out

Object Detection: Owl‐Vit for Videos

Implementation of Google Owl-ViT's model for zero-shot object detection in videos using natural language prompts.

Check it Out

Contribution to joliGEN

I integrated edge detection methods into joliGEN (from ControlNET). I also added Segment-Anything Model from Meta for precise masking.

Check it Out

Multilingual Machine Reading

Development of a multilingual question answering system in English, Finnish and Japanese.

Check it Out

Killian Steunou

Welcome to My Portfolio.

About Me

Professional Experiences

Incoming Intern - AI Research - Idemia

Intern - AI Research - CLS

Intern - Machine Learning Engineer - JoliBrain

Intern - Software Developer (R Shiny) - French Ministry of Agriculture

Academic Background

Master 2 Mathématiques, Vision, Apprentissage - ENS Paris-Saclay

Master 1 Applied Mathematics and Statistics - Toulouse School of Economics

Gap Year - University of Copenhagen

Double Bachelor: Applied Mathematics and Economics - Toulouse School of Economics

Some of my Recent Projects

Video Background Removal

Object Detection: Owl‐Vit for Videos

Contribution to joliGEN

Multilingual Machine Reading

Research

About the Geometric Sensitivity of Adaptative Optimizers to Neural Networks' Structure

Get In Touch With Me