Sayam Kumar
Senior AI Research Engineer at Phaidra
London, England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolSayam Kumar is a Senior AI Research Engineer based in London with 7 years of experience building AI-powered control systems for industrial processes and applied research across healthcare and e‑commerce domains. He combines practical production work at Phaidra with deep probabilistic modeling expertise demonstrated by significant open-source contributions to the widely used PyMC project, including new distributions, improved sampling, and variational inference interfaces. Sayam’s background includes implementing Mean Field and Full Rank ADVI during Google Summer of Code and expanding time-series and VI capabilities in experimental PyMC4, signaling strength in Bayesian methods and TensorFlow Probability. He brings a researcher’s rigor to engineering problems, focusing on robust statistical behavior, edge-case handling, and clear documentation to make advanced probabilistic tools reliable in production.
7 years of coding experience
5 years of employment as a software developer
High School, High School at BCM Arya Model Senior Secondary School, LudhianaBachelor of Technology - BTech, Computer Science, Bachelor of Technology - BTech, Computer Science at Indian Institute of Information Technology, SriCity
English, Hindi, Punjabi
Stackoverflow
31reputation
496reached
0answers
2questions
Github Skills (23)
probabilistic-programming10
variational-inference10
python10
data-modeling10
statistics10
time-series10
tensorflow-probability10
statistic10
numpy10
probabilistic-reasoning10
mcmc10
tensorflow10
statistical-models10
probabilistic-models10
bayesian-inference10
Programming languages (8)
Github contributions (5)
pymc-devs/pymc4
Jan 2020 - Oct 2020
Experimental PyMC interface for TensorFlow Probability. Official work on this project has been discontinued.
Role in this project:
Data Scientist
Data ScientistContributions:5 reviews, 9 commits, 9 PRs in 9 months
Contributions summary:Sayam made significant contributions to the `pymc4` repository, which is an experimental interface for TensorFlow Probability. The user added the Autoregressive distribution, expanding the library's time series capabilities. They also added variational inference interfaces, demonstrating expertise in probabilistic modeling, and included log-likelihood calculations for samples, enhancing the analytical capabilities of the package. Additionally, they addressed minor bugs and made various documentation updates, improving code readability and usability.
pythonpymcmachine-learningdiscontinuedprobability
pymc-devs/pymc
Sep 2020 - Oct 2021
Bayesian Modeling and Probabilistic Programming in Python
Role in this project:Data Scientist
Data ScientistContributions:105 reviews, 18 commits, 27 PRs in 1 year 1 month
Contributions summary:Sayam primarily contributed to the PyMC3 library by addressing issues related to various probability distributions. Their contributions include fixing bugs in existing distributions, adding new distributions like MixtureSameFamily, and improving the random sampling methods for these distributions. They also focused on ensuring the shape and broadcasting behavior of the random samples, and included comprehensive tests to validate the changes. The user also handled edge cases and added examples to clarify how the distributions behave.
pythonbayesian-inferencestatistical-inferencemachine-learningprobabilistic-programming
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.