Hugh Salimbeni
Scientific Director at G-Research
England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolHugh Salimbeni is a Scientific Director at G-Research with a decade of experience applying statistical machine learning and Gaussian process methods to real-world quantitative problems. He progressed from PhD research at Imperial College to senior quant roles, blending rigorous probabilistic modelling with production-focused engineering. His open-source contributions to GPflow—improving variational Gaussian process implementations and testing equivalence across models—reflect a commitment to robust, reproducible ML tooling. Comfortable leading research teams, he pairs deep theoretical training (PhD, MPhil) with practical delivery at the intersection of academia and finance. An early career in maths education hints at strong communication and mentoring skills that complement his technical leadership.
10 years of coding experience
7 years of employment as a software developer
Doctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at Imperial College LondonMaster of Philosophy - MPhil Computational Biology, Master of Philosophy - MPhil Computational Biology at University of Cambridge
Github Skills (12)
machine-learning10
variational-inference10
tensorflow10
python10
gpflow10
gaussian-processes10
bayesian9
bayesian-statistics9
deep-learning8
unit-test7
testing7
unit-testing7
Programming languages (4)
Github contributions (5)
GPflow/GPflow
Nov 2016 - Aug 2019
Gaussian processes in TensorFlow
Role in this project:
ML Engineer
ML EngineerContributions:48 commits, 15 PRs, 52 pushes in 2 years 9 months
Contributions summary:Hugh primarily contributed to the Gaussian processes framework, specifically focusing on variational inference methods. Their commits involved refactoring and improving the Variational Gaussian Process (VGP) implementation, introducing and testing new methods like VGP_opper_archambeau, and resolving dimension issues related to likelihood calculations. They also added tests to ensure the correctness and equivalence of different model implementations.
information-theorygpflowdeep-learningmachine-learningmarkov-chain-monte-carlo
hughsalimbeni/bayesian_benchmarks
May 2018 - Aug 2019
A community repository for benchmarking Bayesian methods
Contributions:103 commits, 15 PRs, 79 pushes in 1 year 2 months
benchmarkingbayesian-inferencedensity-estimationbayesiansampling
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.