Manu Joseph V
Data Scientist
Bengaluru, Karnataka, India
Summary
🤩
RockstarManu Joseph V is a Staff Data Scientist based in Bengaluru with 11 years of experience building production-ready ML solutions for large marketplaces. At Walmart Marketplace he combines rigorous time-series forecasting and tabular deep learning expertise to drive data-driven decisions at scale. His open-source work includes foundational contributions to a Packt time-series forecasting repository and significant engineering on a PyTorch Tabular framework—highlighting strengths in reproducible notebooks, feature engineering, experiment management, and model optimization. Comfortable moving between research-style prototyping and robust engineering, he repeatedly restructures and hardens codebases to make models reliable in production. Colleagues would note his pragmatic focus on practical examples and the way he surfaces subtle data-generation and encoding issues early in projects.
11 years of coding experience
Github Skills (13)
pandas10
pytorch10
jupyter-notebook10
machine-learning10
time-series10
pytorch-lightning10
deep-learning10
feature-engineering10
datatable10
python10
tabular10
data-analysis10
numpy9
Programming languages (5)
Github contributions (5)
PacktPublishing/Modern-Time-Series-Forecasting-with-Python
Aug 2021 - Jan 2023
Modern Time Series Forecasting with Python, published by Packt
Role in this project:
Data Scientist
Data ScientistContributions:37 commits, 4 PRs, 38 pushes in 1 year 5 months
Contributions summary:Manu's initial commits established the project's foundation, adding the first chapter with notebooks demonstrating Data Generating Processes and Synthetic Time Series. The user then focused on removing and restructuring the notebook files to change the folder structure. The user then added multiple notebooks demonstrating and implementing chapters 1-6 to the project. This suggests a focus on implementing the book's content with code examples.
manujosephv/pytorch_tabular
Dec 2020 - Jan 2023
A standard framework for modelling Deep Learning Models for tabular data
Role in this project:ML Engineer
ML EngineerContributions:6 releases, 90 reviews, 320 commits in 2 years 1 month
Contributions summary:Manu's commits focus on implementing and integrating machine learning models for tabular data. They made several modifications to the core model, including fixes for the experiment version manager, trainer, and metric problems. They also incorporated a new category encoder, highlighting their work on feature engineering and model optimization within the PyTorch Tabular framework. Furthermore, they worked with the codebase of the model to test the model, and also adapted the model for classification.
pytorchdeep-learningtabular-datamachine-learningdeep-learning-models
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.