experience/
├── opalin/
│   ├── role: Founding Engineer, Embodied AI
│   ├── period: Feb 2026 - Present
│   ├── location: Redwood City, CA
│   ├── desc: building robots at Opalin, deploying them right now to augment factory workers
│   └── tasks/
│       ├── Teleop-to-fleet pipeline: multi-teleoperator data collection, filtering & evaluation within a day
│       ├── VLA & WAM: architectures, training, fine-tuning & real-time evaluation of visuomotor policies
│       ├── Data curation: 3D dataset viewer, per-task statistics & outlier detection (50% -> 98% success with half the data)
│       └── Real-time arm control: kinematics & dynamics, impedance control (0.1 mm)
├── thales-las/
│   ├── role: Embedded AI Software Engineer
│   ├── period: Sept 2024 - Feb 2025
│   ├── location: Vélizy-Villacoublay, France
│   └── tasks/
│       ├── Decentralized embedded AI for drone swarm coordination & control
│       ├── Embedded Lua scripting engine & on-device LLM voice-to-script generation (98% correctness)
│       ├── Cut new-mission deployment time by 10x and improved reliability by 4x
│       └── Part of Thales' COHESION demonstrator (JDEC 2024), featured in French media
├── cnrs/
│   ├── role: Research Assistant - Autonomous Vehicle Control Lead
│   ├── period: Feb 2024 - Jun 2025
│   ├── location: Compiègne, France
│   └── tasks/
│       ├── Waypoint-based control architecture for shared autonomous vehicles in dense urban environments
│       ├── Sim-to-real deployment onto a Renault Zoe (ROS/Python): waypoint control, adaptive cruise control, obstacle avoidance (99% safety in sim)
│       └── Won 1st School Award (2024) and 1st place Open Category (2025), UTAC Challenge 🥇
└── junior-utc/
    ├── role: Business Manager
    ├── period: Sept 2023 - Sept 2024
    ├── location: Compiègne, France
    └── desc: Led technical projects for Airbus, €30k+ in projects

waterloo_ms_ai_2025.txt    utc_ms_ce_2021-2026.txt

University of Waterloo, Canada
M.Sc. Artificial Intelligence — Fall 2025
Deep Learning, Machine Learning, Image Processing. GPA 4.0/4.0
Professors: Bryan Tripp, Terrence C. Stewart

Université de Technologie de Compiègne, France
M.Sc. Computer Engineering — 2021 - 2026
Embedded & Autonomous Systems: Robotics Control, Embedded Systems, Kinematics. GPA 5.0/5.0

📄 Publications
- Behavior Cloning of MPC for 3-DOF Robotic Manipulators (arXiv:2606.00383)
  Poster at the Workshop on RL in the Era of Imitation Learning @ IEEE ICRA 2026
  Distilled an MPC policy into neural surrogates: 3x lower inference latency at 85% success

🛠️ Open Source
- fathom: zero-overhead Rust visualization of multimodal spatiotemporal data (video, depth, point clouds, trajectories)
- SuperMimic reproduction: differentiable world model for humanoid whole-body control, trained via BPTT over imagined rollouts
- Reachy Mini & SO-ARM-101 manipulation stacks; peillute (distributed P2P payments in Rust); Kaggle Brain-to-Text '25

🚀 Hackathons
- UTAC Autonomous vehicle challenge (France) : best school award 2024 | won 2025 edition
- FIT Coding Challenge 2025 (Bosnia)
- UTC x st2i hackathon (France) : winners
- SWARM, The World's Largest Swarm Robotics Hackathon (Canada)
- Mistral AI Worldwide Hackathon 2026: build the next era of AI (San Francisco)
- Seeed x NVIDIA x Hugging Face 2026 Embodied AI Hackathon (Santa Clara) : won 2nd prize
- Open World Hackathon 2026 by VLGE (San Francisco) : won 2nd prize

On branch main
Your branch is up to date with 'origin/main'.

Changes to be committed:
  (use "git restore --staged <file>..." to unstage)
        new file:   skills/vla-models
        new file:   skills/robotic-control
        new file:   skills/embedded-systems
        modified:   ambition.txt
        modified:   caffeine-level.txt

Untracked files:
  (use "git add <file>..." to include in what will be committed)
        next-big-thing/