Internship: ✅ May–August 2026 at Manako (team behind SN44/Score on Bittensor)
Master's Program: Open to work-study alternance for 2026–2028
Focus: Computer vision, ML systems, Web3, and real-time applications
Education: Year 2 @ Holberton School — RNCP Level 6
I build practical AI/ML tools with a focus on computer vision, automation, and real-time systems. I care about shipping projects that are understandable, testable, and documented well enough for other developers to reuse.
What I spend time on now:
- 🤖 Machine Learning & AI — Computer vision, model training, and deployment-oriented experiments
- ⛓️ Web3 & real-time systems — Decentralized experiments, WebSocket-based architecture, backend orchestration, and state sync
- 🏠 Home Lab — Hosting VoxelPlace services and testing new infrastructure setups
📌 Recent Activity Log (Click to expand)
Automatically updated from my public GitHub activity.
- Apr 29, 2026 — Watch warpdotdev/warp
- Apr 29, 2026 — ⭐ Starred warpdotdev/warp
- Apr 28, 2026 — Watch Facepunch/sbox-public
- Apr 28, 2026 — ⭐ Starred Facepunch/sbox-public
- Apr 27, 2026 — 📤 Pushed 🔥 MaKSiiMe/holbertonschool-interview
- Apr 25, 2026 — 📤 Pushed 🔥 MaKSiiMe/VoxelPlace
| What I build | Main tools |
|---|---|
| Computer vision & ML experiments | Python, PyTorch, TensorFlow, OpenCV, Scikit-learn |
| Real-time backends & APIs | FastAPI, Flask, Fastify, Redis, Socket.io |
| Interactive web apps | React, Next.js, TypeScript, PixiJS |
| Data, infra & operations | PostgreSQL, MongoDB, Docker, Linux, Git |
| Systems & scripting fundamentals | C, Bash/Shell, JavaScript |
- Master's alternance (2026-2028) in AI/ML, computer vision, or applied distributed systems.
- Mission types: model prototyping, real-time backend engineering, and production-oriented experimentation.
- Work setup: South of France preferred, open to hybrid/remote depending on the team.
| Project | Description | Stack |
|---|---|---|
| Pokémon Blue AI Agent | Modular RL + CV agent for Pokémon Blue, combining PPO/DQN with YOLO for autonomous navigation and battle decisions. Result: reached and beat the first Champion autonomously. | Python PyTorch YOLO OpenAI Gym PyBoy |
| VoxelPlace | Real-time collaborative pixel canvas playable from browser and Minecraft, with synchronized state across clients and server. Result: one shared canvas synchronized across 2 client environments (web + Minecraft). | Next.js Fastify Socket.io Redis TypeScript Java |
- AI/ML Foundations: Linear algebra, calculus, CNNs, object detection, and model evaluation.
- Applied Engineering: Building end-to-end projects that combine ML, Web3 components, and real-time backend systems.
- Community & Events: Chiliz Hackathon (Sport x Blockchain), Proof of Talk, VivaTech.