I recently graduated Summa Cum Laude from UCLouvain (École Polytechnique de Louvain) with an MSc in Computer Science & Engineering, specializing in Artificial Intelligence.

I'm looking to join a team as an AI / ML Engineer, building AI systems that actually ship to production.

I'm passionate about Computer Vision, Natural Language Processing, and Generative AI (LLMs, RAG & Agentic AI) — and I enjoy engineering the entire pipeline behind them, from large-scale training on HPC GPU clusters (SLURM) to Docker-containerized deployment on Azure and full-stack delivery.

I'm open to AI / ML Engineer roles in Belgium (on-site/hybrid) or fully remote across the EU.

Reach out via LinkedIn or at mathis.delsart@gmail.com.

Deep Reinforcement Learning for Competitive Agents in MicroRTS — Architecture, Training, and Tournament Evaluation

UCLouvain · Completed June 2026 · Accepted at IEEE CoG 2026

A DRL agent for real-time strategy games, fusing a U-Net spatial encoder with an entity-level Transformer (UECD) and trained end-to-end with a modular PPO pipeline on HPC GPU clusters (SLURM) — every design decision individually ablated.

• 🥇 96.67% pool win rate — tops a 19-agent tournament (1st on 4 of 5 metrics)
• ⚔️ Beats RAISocketAI (CoG competition winner) in 65.7% of head-to-heads
• ⚡ Trained on just 9.47 GPU-days — vs the 23.6 reported by RAISocketAI

_{UECD-Best (ours, top-left) vs RAISocketAI (bottom-right)}

📂 View the repository →

Deep Learning & Computer Vision

Project	Repo	Description	Stack
Satellite-Image-Segmentation		Satellite image segmentation with supervised & self-supervised learning
3D-Hand-Gesture-Classification		Comparative study: KNN, LSTM, DenseNet, HMM for 3D gesture recognition
GANs-Diffusion-Kernels		Advanced generative models: GANs, diffusion models, kernel methods
Heart-Failure-Detection-Smurf		Heart-failure risk prediction on a fictional "Smurf Society" dataset (synthetic, not real clinical data)
ML-Classification		Supervised learning models for medical & biological datasets
Celebrity-Dataset-Explorer		Interactive face attribute analysis with clustering & dimensionality reduction

Natural Language Processing & LLMs

Project	Repo	Description	Stack
Sourcio — RAG Study Assistant		Deployed RAG tutor answering only from your own courses — citation-by-construction (the model never sees a page number) + a three-layer refusal guard. Live app: 100% citations, 94% refusal accuracy
NLP-Models		Progressive NLP pipeline: Question Classification (Naive Bayes), Vector Semantics (TF-IDF/PPMI/Word2Vec), BERT-based QA (SQuAD)
Text-Prediction-TwitOZ		Intelligent text prediction system using N-grams algorithm with real-time word suggestions

Game AI & Reinforcement Learning

Project	Repo	Description	Stack
Snakes-Ladders-MDP		Optimal strategy via Markov Decision Processes & Q-Learning
AI-Odyssey		Search algorithms: uninformed to adversarial game-playing agents
AI-Connect4-Game		Connect 4 with Minimax AI — Can you beat a machine that thinks 4 moves ahead?

Data Science & Analytics

Project	Repo	Description	Stack
Data-Science-Portfolio		Applied statistics & data science: energy analysis, star classification, species clustering
Pattern-Mining		Pattern mining: frequent itemsets, sequential patterns, and Bayesian networks
Database-SQL		SQL query optimization and Neo4j graph analysis with Cypher

Systems Programming & Compilers

Project	Repo	Description	Stack
Custom-Compiler		Full compiler: custom language → JVM bytecode
Prolog-Python-Analyzer		Prolog-powered static analyzer for Python code detecting 13+ code smells through declarative pattern matching on ASTs
Scheme-OOP		Building OOP from scratch in pure Scheme: classes, inheritance, and polymorphism from closures and message passing
Java-COP		Context-Oriented Programming in Java using reflection, dynamic proxies, and bytecode instrumentation with custom profiler agent

Low-Level Systems & Embedded

Project	Repo	Description	Stack
CORDIC-FPGA-Accelerator		High-performance FPGA accelerator integrated with ARM processor
High-Performance-Server		High-performance HTTP matrix multiplication server on NGINX with progressive optimizations: cache-aware algorithms, SIMD (AVX-512), and CUDA GPU acceleration
Multithreaded-Bellman-Ford		Parallel single-source shortest-path (Bellman-Ford) in C, parallelized with a POSIX-thread pool — ~6× faster on large graphs
Dynamic-Memory-Allocator		Custom heap allocator with next-fit & coalescing
Concurrent-Bench-Sync-Primitives		Performance benchmarking framework for custom synchronization primitives (TS, TTS, BTTS) vs POSIX on Reader-Writer, Dining Philosophers, Producer-Consumer
Tar-Archive-Parser		Lightweight TAR archive parser in C — validation, file operations, symlink resolution

Numerical Computing & Optimization

Project	Repo	Description	Stack
FEM-Bridge-Simulator		Finite element simulator for bridge structures under static loads, featuring 2D elasticity modeling and animations
Optimization-Algorithm		Solving NP-hard problems: exact methods, relaxations, heuristics
Algorithm-and-Data-Structures		Clean and well-tested core data structures and algorithms with performance focus

Web & Mobile Development

Project	Repo	Description	Stack
Shopping-Cart-Application		Scalable microservices e-commerce platform with custom recommendations, smart ranking, and dynamic shopping
Private-Tutoring		Bilingual, responsive tutoring website, statically deployed to GitHub Pages
Energy-Tracker-MeterBuddy		Android energy-consumption tracker with charts, Room storage, and Firebase sync

Games

Project	Repo	Description	Stack
Chess-Game		Full chess rules, multiple themes, and a clean Pygame interface
Flappy-Bird-Game		Classic Flappy Bird recreation with physics and animations
Modern-Snake-Game		Modern, polished Snake game with gradients and animations

Coursework & Foundations

Project	Repo	Description	Stack
EPL-Chronicles		Aggregated UCLouvain coursework: systems, ML, data science & programming paradigms