Sioni Summers
Applied Physicist at CERN
Meyrin, Geneva, Switzerland
Summary
🤩
Rockstar🎓
Top SchoolSioni Summers is an applied physicist at CERN with a decade of experience building ultra-low-latency data filtering and hardware-aware ML systems for high-energy physics. He bridges physics and software engineering, contributing to CMS offline software and FPGA-targeted ML tooling like hls4ml, where he implemented hard-coded decision trees in C++ and adapted models for quantized, FPGA deployment. His background includes a PhD from Imperial College London and roles as Senior Fellow and Research Associate, giving him deep domain knowledge in detector-level data processing and system integration. Known for pragmatic, production-focused contributions, he often turns complex algorithmic ideas into efficient firmware and backend code that meets strict latency and resource constraints.
10 years of coding experience
1 year of employment as a software developer
PhD, High Energy Physics, PhD, High Energy Physics at Imperial College London
Github Skills (16)
fpga10
mask-rcnn10
keras10
faster-rcnn10
machine-learning10
jupyter-notebook10
c-language10
tensorflow10
cprogramming-language10
python10
back-end-development10
data-structure9
cer9
data-structures9
web-framework9
Programming languages (6)
Github contributions (5)
fastmachinelearning/hls4ml-tutorial
Jun 2020 - Jul 2022
Tutorial notebooks for hls4ml
Role in this project:
ML Engineer
ML EngineerContributions:1 review, 52 commits, 17 PRs in 2 years 1 month
Contributions summary:Sioni's commits primarily involve updating and modifying Jupyter notebooks related to the hls4ml tutorial. The changes include modifications to code within the notebooks for training, configuring, and deploying machine learning models, particularly CNNs, boosted decision trees (BDTs), and quantized models. They demonstrate proficiency in applying hls4ml and conifer libraries, along with the creation of bitfiles for FPGA deployment on a PYNQ-Z2 board, aligning with core tutorial objectives.
jupyter-notebooknotebooksjupytervoilabinder
fastmachinelearning/hls4ml
May 2018 - Oct 2022
Machine learning on FPGAs using HLS
Role in this project:ML Engineer
ML EngineerContributions:6 releases, 148 reviews, 272 commits in 4 years 6 months
Contributions summary:Sioni implemented and tested a hard-coded decision tree implementation in C++ for the hls4ml project. This involved creating functions to represent the decision tree logic and verifying its functionality. The user further contributed by integrating and merging new branches, and incorporating new features such as support for pooling layers and adjustments for QKeras quantizers.
pytorchpythonvivadofpgasonnx
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.