David Silva
Robotics And Perception Engineer at Stealth
Germany
Summary
🤩
Rockstar🎓
Top SchoolDavid Silva is a Robotics and Perception Engineer with a decade of hands-on experience building AI systems for autonomous vehicles, drones, and industrial robots across Europe. He combines deep ML and computer-vision expertise—PyTorch model development, custom loss design, IoU evaluation tooling and learning-rate utilities—with production deployment skills in C++, TensorRT, CUDA and ROS for Jetson-class platforms. At WINGCOPTER and Aptiv he shipped detect-and-avoid, landing-spot, and multi-radar tracking solutions, and he has automated multi-GPU training workflows and maintained self-hosted CI runners. His open-source work includes practical tooling (PyTorch ENet implementation and a packaged LR finder) that emphasizes model evaluation and reproducible experiments. Based in Germany and trained as a mechanical engineer, he bridges algorithm research and embedded systems engineering to deliver performant perception stacks for safety-critical robotics.
10 years of coding experience
8 years of employment as a software developer
Master’s Degree Mechanical Engineering, Master’s Degree Mechanical Engineering at Universidade de Aveiro
Portuguese, English, German
Github Skills (9)
computer-vision10
pytorch10
machine-learning10
python10
metric10
deep-learning9
pip9
matplotlib7
numpy6
Programming languages (4)
Github contributions (5)
davidtvs/pytorch-lr-finder
Nov 2018 - Sep 2020
A learning rate range test implementation in PyTorch
Role in this project:
ML Engineer
ML EngineerContributions:1 release, 7 reviews, 52 commits in 1 year 10 months
Contributions summary:David contributed to the development of a PyTorch-based learning rate finder. They enabled the restoration of model and optimizer states using a reset function, improving the usability of the tool. Furthermore, they packaged the project as a pip package, making it easier for others to install and use. The commits also included code format improvements and fixes to documentation, ensuring code quality and clarity.
pytorchrangelearning-rate
davidtvs/PyTorch-ENet
Feb 2018 - May 2021
PyTorch implementation of ENet
Role in this project:ML Engineer
ML EngineerContributions:2 reviews, 71 commits, 4 PRs in 3 years 3 months
Contributions summary:David implemented and added essential components for evaluating the ENet model. Their contributions focused on creating classes for IoU metric calculation, including handling multi-label confusion matrices. The user also developed a validation class and integrated the IoU metric into the main training script, demonstrating a focus on model evaluation and performance analysis. The user's work directly supported the core functionality of the ENet PyTorch implementation.
pytorchdeep-learningenetpytorch-implementation
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.