Sven Chilton
Deep Learning Developer Advocate
Berkeley, California, United States
Summary
🤩
Rockstar🎓
Top SchoolSven Chilton is a Deep Learning Developer Advocate at NVIDIA with 11 years of experience translating advanced AI and computational physics into practical developer tools, tutorials, and courses. He builds and documents end-to-end workflows for speech AI, NLP, and computer vision—authoring Jupyter notebooks, demo videos, and NVIDIA DLI curricula—and helps customers integrate SDKs like Riva and Torch-TensorRT. With a PhD in computational electrodynamics and a strong software toolbox (C++, Python, R, MATLAB, Fortran), he uniquely bridges rigorous numerical methods and production ML optimization on GPUs. Sven contributes to high-profile open-source projects such as PyTorch/TensorRT, including hands-on work optimizing Hugging Face BERT models, and has taught and mentored learners across DeepLearning.AI and FourthBrain. He’s driven by making AI accessible and deployable, whether enabling speech on a Boston Dynamics Spot or shrinking inference footprints with TensorRT.
11 years of coding experience
12 years of employment as a software developer
Master’s Degree, Nuclear Engineering, GPA 3.9/4.0, Master’s Degree, Nuclear Engineering, GPA 3.9/4.0 at University of California, BerkeleyBachelor's Degree, Physics, GPA 4.5/5.0, Bachelor's Degree, Physics, GPA 4.5/5.0 at Massachusetts Institute of Technology
English, Swedish, German
Github Skills (11)
transformers10
pytorch10
machine-learning10
deeplearning-ai10
huggingface10
deep-learning10
optimisation10
bert10
optimization10
python9
cuda9
Programming languages (2)
Github contributions (5)
pytorch/TensorRT
Jan 2022 - Jul 2022
PyTorch/TorchScript/FX compiler for NVIDIA GPUs using TensorRT
Role in this project:
ML Engineer
ML EngineerContributions:2 reviews, 11 commits, 11 PRs in 6 months
Contributions summary:Sven contributed to the `pytorch/tensorrt` repository by modifying and adding content related to running Hugging Face BERT models. The user added a new notebook showcasing how to optimize a Hugging Face BERT transformer (bert-base-uncased) for masked language modeling. Furthermore, the user changed the Torch-TRT logging level to avoid long printouts and added comments.
cudapytorchjetsonnvidiadeep-learning
VahidooX/tutorials
Jan 2023 - Jan 2023
NVIDIA Riva runnable tutorials
Contributions:46 pushes in 1 day
nvidiarivarunnable
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.