Juan Da Silva

Juan Da Silva

Mathematician & Data Scientist · Master's Student at TUM
Deep Learning · 3D Vision · Visual Recognition · Generative Models · CUDA & GPU Optimization

I am a Master's student in the Mathematics in Data Science program at the Technical University of Munich (TUM). My academic interests lie at the intersection of applied mathematics, machine learning, and mathematical computer vision, focusing on developing robust geometric and analytical frameworks for data analysis.


Positions

🛰️
Graduate Research Student
Incoming 10/2026 — Present
Airbus · Digital Payload Processors Team
Research and development of enhanced AI models for autonomous Fault Detection, Isolation, and Recovery (FDIR) in spacecraft digital payload processors.
🧠
Graduate Research Student
Current 04/2026 — Present
Working in the group of Prof. Daniel Cremers under the supervision of Dr. Héctor Andrade Loarca on hardware-aware neural rendering, CUDA kernel optimization, and frequency-domain representations for text-to-3D generation.

Working Papers


About

I am a Master's student in the Mathematics in Data Science program at the Technical University of Munich (TUM), where I also work as a Graduate Research Student in the Computer Vision Group under Prof. Daniel Cremers and Dr. Héctor Andrade Loarca. Previously, I completed double Bachelor's degrees in Mathematics and Computer Science at Heidelberg University.

My research interests sit at the intersection of mathematical computer vision, deep learning theory, and scientific computing. I am particularly focused on hardware-aware neural rendering, generative 3D modeling, low-level GPU acceleration (CUDA), and the mathematical foundations of deep learning.

Research
3D Vision Visual Recognition Neural Rendering Generative 3D Modeling CUDA & GPU Computing Deep Learning Theory
Stack
Python C/C++ CUDA MATLAB Java R SQL PyTorch JAX TensorFlow Hugging Face OpenCV Git Docker
Languages
Spanish (Native) Portuguese (Native) English (C1) German (C1) Italian (B1)