Yashu Seth
Senior Machine Learning Engineer at Uber
Bengaluru, Karnataka, India
Summary
👤
Senior🎓
Top SchoolYashu Seth is a Senior Machine Learning Engineer with 10 years of experience building production-grade AI systems and MLOps pipelines, currently driving ML at Uber from Bengaluru. He has led impactful product initiatives at LinkedIn—designing a content-to-group recommendation system with a 67% site-wide impact and generative QA for groups (both patent-pending)—and delivered patented document extraction and anonymization solutions at Ushur. Skilled across probabilistic models, C++ ML libraries and backend engineering, he contributes to prominent open-source projects like pgmpy and mlpack, improving core Bayesian network validation and distribution training methods. Yashu combines rigorous QA and testing discipline with system architecture experience (RabbitMQ/ZeroMQ, MLflow, Docker), enabling reliable model deployment at scale. Notably, his early open-source work adding Gaussian distributions to pgmpy under Google Summer of Code signals a long-standing commitment to robust probabilistic modeling.
10 years of coding experience
8 years of employment as a software developer
B.Tech. Electronics Engineering, B.Tech. Electronics Engineering at Indian Institute of Technology (Banaras Hindu University), Varanasi
English, Hindi
Stackoverflow
91reputation
11kreached
0answers
3questions
Github Skills (19)
unit-testing10
type-check10
python10
testing10
machine-learning10
cplus10
bayesian-network10
type-checking10
cpp10
pytest9
regression9
perceptron8
msys6
mingw6
gcc6
Programming languages (3)
Github contributions (5)
pgmpy/pgmpy
Sep 2015 - Jan 2017
Python Library for learning (Structure and Parameter), inference (Probabilistic and Causal), and simulations in Bayesian Networks.
Role in this project:
Back-end Developer & QA Engineer
Back-end Developer & QA EngineerContributions:201 commits, 40 PRs, 546 comments in 1 year 4 months
Contributions summary:Yashu primarily focused on improving the quality and robustness of the `pgmpy` library. Their contributions involved adding type checking to factor operations and comprehensive unit tests for various edge cases, including those that were prone to TypeError exceptions. They also fixed a bug related to the `check_model` method in the BayesianModel and DynamicBayesianNetwork classes, which suggests a focus on the core model validation logic. Additionally, the user corrected documentation typos.
causal-modelspythondagbayesian-inferencecausal
mlpack/mlpack
Dec 2016 - Jan 2017
mlpack: a fast, header-only C++ machine learning library
Role in this project:ML Engineer / QA Engineer
ML Engineer / QA EngineerContributions:10 commits, 2 PRs, 18 comments in 15 days
Contributions summary:Yashu contributed to the `mlpack` library by implementing a new training method for the `GammaDistribution` class, enabling parameter estimation with observation probabilities. They then added multiple test cases to ensure the correctness and robustness of the newly implemented `Train` function. Further contributions include adding their name to the core contributors list, and correcting a typo in the documentation of the Perceptron method. Additional tests for the `SimpleWeightUpdate` policy were added as well.
regressionheaderdeep-learningscientific-computingc-plus-plus
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.