Oliver Holworthy
Senior Machine Learning Engineer at NVIDIA
London, England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolOliver Holworthy is a Senior Machine Learning Engineer based in London with 11 years of experience building production-grade ML systems and MLOps infrastructure. Currently at NVIDIA, he focuses on making large-scale recommender tooling more reliable and deployable, contributing to high-profile open-source projects like NVIDIA Merlin, Transformers4Rec and NVTabular. He brings a strong DevOps sensibility to ML engineering—refactoring build pipelines, container images, tests and type checks to improve maintainability and reproducible deployments. Previously he spent years engineering data-driven features and services at Lyst and applying quantitative computing in computational finance, giving him a rare blend of product-focused ML and robust backend engineering. Colleagues rely on him to bridge research-grade models and production constraints, often surfacing pragmatic fixes (pinning dependencies, aligning data types) that prevent subtle failures at scale.
11 years of coding experience
6 years of employment as a software developer
MSci, MSci at Imperial College LondonKing Edward VI School, Southampton
Github Skills (32)
github-ci10
recommendation-system10
docker10
preprocessing10
python10
preprocess10
mypy10
machine-learning10
feature-engineering10
dockers10
deep-learning10
tensorflow10
precommit10
pre-commit-hook10
githubaction-workflow10
Programming languages (4)
Github contributions (5)
NVIDIA-Merlin/Merlin
May 2022 - Dec 2022
NVIDIA Merlin is an open source library providing end-to-end GPU-accelerated recommender systems, from feature engineering and preprocessing to training deep learning models and running inference in production.
Role in this project:
MLOps Engineer
MLOps EngineerContributions:42 reviews, 13 commits, 35 PRs in 6 months
Contributions summary:Oliver's contributions primarily focused on improving the Merlin framework's deployment and testing infrastructure. This includes refactoring Docker images to remove unnecessary entrypoints, streamlining build processes, and adjusting test timeouts to accommodate larger model training and deployment procedures. The user also made changes to example notebooks and integration tests, updating the build process to pin protobuf versions, and aligning data types for compatibility. These changes collectively enhance the framework's reliability and efficiency for deploying machine learning models.
in-productiontrainingdeep-learning-modelsengineeringtensorflow
NVIDIA-Merlin/NVTabular
Aug 2022 - Jan 2023
NVTabular is a feature engineering and preprocessing library for tabular data designed to quickly and easily manipulate terabyte scale datasets used to train deep learning based recommender systems.
Role in this project:ML Engineer
ML EngineerContributions:59 reviews, 10 commits, 55 PRs in 5 months
Contributions summary:Oliver made several contributions related to the NVTabular library, including updating the versioneer package, modifying the Categorify and DropLowCardinality operators, and adjusting the schema of value counts. They also handled the data loader as an iterator, and updated package requirements by removing unnecessary extras. The user implemented tests for value counts and addressed issues with sparse_max.
tsneengineeringtensorflowpreprocessingnvidia
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.