Hessam Mehr
Lecturer In Chemistry And AI For Science at King's College London
London, England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolHessam Mehr is a Lecturer in Chemistry and AI for Science with 15 years of interdisciplinary experience bridging supramolecular and synthetic organic chemistry with probabilistic modelling and robotics. His work focuses on AI-first chemical discovery, developing novel experimental regimes and abstractions for digitized, robotic synthesis and automated probabilistic discovery. He has applied signal processing and probabilistic inference to practical problems such as NMR-based mixture identification at Health Canada and led research into uncovering biased "dark" reactions during a Leverhulme Fellowship at the University of Glasgow. An active contributor to probabilistic programming ecosystems, he has improved core functionality in projects like Turing.jl and added advanced distributions and vectorized MCMC enhancements to NumPyro, reflecting a strong back-end and computational-statistics skill set. Trained as a PhD chemist with an undergraduate background in electrical and electronics engineering, he uniquely blends lab-first chemical intuition with language, compilation, and inference techniques for physical processes. Based in London, he is particularly interested in bringing probabilistic/logic methods and machine intelligence to explore previously inaccessible regions of chemical and biomolecular space.
15 years of coding experience
4 years of employment as a software developer
Doctor of Philosophy - PhD Organic Chemistry, Doctor of Philosophy - PhD Organic Chemistry at The University of British ColumbiaBachelor of Science - BS Electrical and Electronics Engineering, Bachelor of Science - BS Electrical and Electronics Engineering at Sharif University of Technology
Github Skills (19)
algorithm10
algorithms10
probabilistic-programming10
python10
inference10
numpy10
mcmc10
bayesian10
jax10
bayesian-inference10
architecture9
testing9
machine-learning9
architectures9
build-automation8
Programming languages (13)
Github contributions (5)
TuringLang/Turing.jl
Apr 2018 - Sep 2018
Bayesian inference with probabilistic programming.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:28 commits, 6 PRs, 4 pushes in 5 months
Contributions summary:Hessam primarily contributed to the build process and core functionalities of the Julia-based probabilistic programming library. This involved migrating the build system to BinaryProvider, updating dependencies, and correcting deprecation warnings within the codebase. Additionally, the user addressed mutability and vectorization issues, and removed unnecessary broadcasts, suggesting a focus on improving code efficiency and compatibility with the Julia language.
julia-languagebayesian-inferencebayesian-methodsbayesian-neural-networkshamiltonian-monte-carlo
pyro-ppl/numpyro
Oct 2021 - Dec 2022
Probabilistic programming with NumPy powered by JAX for autograd and JIT compilation to GPU/TPU/CPU.
Role in this project:ML Engineer
ML EngineerContributions:13 reviews, 5 commits, 8 PRs in 1 year 1 month
Contributions summary:Hessam contributed to the probabilistic programming library `numpyro` by implementing new distributions, improving existing sampling methods, and addressing code quality issues. They added a `MultivariateBeta` distribution, generalized it to a `GaussianCopula`, and implemented several related distributions with associated tests. The user also made improvements to MCMC sampling, specifically around vectorized sampling and shape calculations. Additionally, the user addressed CI failures and implemented code refactoring.
pythonautomatic-differentiationbayesian-inferencetpucpu
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.