Summary
I am an AI researcher and engineer with expertise in deep learning, computer vision, natural language processing and optimization. My experience spans from designing efficient AI inference pipelines to developing scalable web and cloud applications. I have worked with PyTorch, CUDA, ONNX, and Apache TVM to optimize AI models for real-time performance, particularly on embedded and high-performance computing (HPC) systems.
Beyond AI, I have hands-on experience in web development and cloud infrastructure, having built secure and scalable applications using Django, Node.js, Docker, and NGINX. I have deployed and maintained web services on VPS and cloud platforms, ensuring robust and efficient software architectures.
My background in computational science, physics, and applied mathematics allows me to tackle complex problems with a data-driven approach. I thrive in fast-paced, research-driven environments, where I can bridge the gap between cutting-edge AI models and real-world deployment challenges.
• Technical Skills: AI & Deep Learning (PyTorch, TensorRT, ONNX), HPC & CUDA Optimization, Full-Stack Web Development (Django, HTMX, TailwindCSS), Cloud & DevOps (Docker, NGINX, PostgreSQL), Embedded AI.
• Interests: AI for real-time systems, AI on the edge, HPC, multimodal models, and inference optimization.
• Looking for: Opportunities in AI research, ML engineering, and optimization at the intersection of deep learning, cloud computing, and embedded AI.
Let’s connect!
Professional experiences
Software engineer
Stealth Startup
From August 2024 to January 2025
• Designed and implemented the entire software architecture of the startup across multiple servers using Docker.
• Developed and deployed secure web applications on VPS using Django and Node.js.
• Configured NGINX as a reverse proxy to enable HTTPS communication and load balancing.
• Created user-friendly frontends using simple tools such as HTMX, TailwindCSS, and Alpine.js.
Artificial intelligence intern
BeEmotion.ai
From June 2024 to July 2024
• Engineered an AI system capable of processing audio inputs, generating personality‑driven responses, and synthesizing speech.
• Optimized inference pipelines to minimize end‑to‑end latency using advanced tools such as TVM and ONNX.
• Adapted key libraries from x86 to ARM architectures, enabling efficient utilization of embedded CUDA resources.
Student research assistant
EPFL
From April 2024 to May 2024
• Researched multimodal models that integrate vision and language processing to enhance image content analysis.
• Explored the application of these models for visual question answering (VQA) in autonomous driving scenarios.
• Developed custom datasets and fine‑tuned open‑source models, achieving performance gains over GPT‑4.
Technical assistant in 3d modeling, web development and vr
Université de Lausanne
From August 2022 to December 2023
• Developed IT solutions for PhD researchers, reducing software costs by CHF 20K through in‑house alternatives.
• Built and customized VR experiments using Unity, improving data accuracy by 10% with advanced measurement functionalities.
• Created a survey platform using oTree, HTML, and CSS, incorporating anti‑bot mechanisms to cut incoherent data by 17%.
• Developed REST APIs using Django Rest Framework and PostgreSQL for seamless integration between Unity, Django, and React interfaces.
Degree
Languages
Français - Langue maternelle
Anglais - Courant
Allemand - Technique
linkedin