Pradnyesh Joshi
Applied Scientist II at Microsoft
Seattle, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolPradnyesh Joshi is an Applied Scientist II at Microsoft with eleven years of experience building end-to-end machine learning solutions, currently focusing on recommendations for the gaming industry within Azure Gaming Services. He combines strong statistical foundations (MS Statistics, UIUC, 4.0 GPA) with practical ML engineering—deploying recommenders, DeepFM/BPR systems, and triplet-loss embedding models in production on GCP and Azure. His background spans NLP, safety- and pricing-related ML in enterprise settings, and refactoring data platforms into scalable Python microservices with PySpark. An active open-source contributor, he has improved Microsoft’s popular recommenders repo by debugging and tightening RBM example implementations, reflecting attention to reproducibility and model correctness. Pragmatic and product-minded, he also builds serving layers and client-facing UI tweaks (Node/TypeScript, React) to close the loop from model to user experience. Based in Seattle, he brings a mix of research rigor and hands-on deployment experience that accelerates ML-driven personalization.
11 years of coding experience
4 years of employment as a software developer
University of Illinois Urbana-ChampaignBachelor of Technology, Information Technology, Bachelor of Technology, Information Technology at National Institute of Technology Karnataka
English, Hindi, Marathi
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (6)
jupyter-notebook10
machine-learning10
recommendation-system10
python10
data-science9
deep-learning9
Programming languages (2)
Github contributions (5)
recommenders-team/recommenders
Dec 2021 - Jul 2022
Best Practices on Recommendation Systems
Role in this project:
Data Scientist
Data ScientistContributions:49 reviews, 304 commits, 45 PRs in 7 months
Contributions summary:Pradnyesh fixed an issue in the `gen_affinity_matrix` function by correcting its return type. The code changes primarily involve modifications to a Jupyter notebook example related to Restricted Boltzmann Machines (RBMs), including minor adjustments in the notebook, rerunning the notebook, and fixing issues related to the model's parameters and outputs. These contributions suggest a focus on enhancing and debugging the RBM model implementation within the recommendation system context.
recommendation-systemspythonjupyter-notebookoperationalizationmicrosoft
pradnyeshjoshi/gender_difference_analysis
Oct 2020 - Jul 2021
Contributions:10 pushes, 3 branches in 9 months
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.