Prateek Desai
Machine Learning Engineer 2 at Amazon Web Services (AWS)
New York, New York, United States
Summary
🤩
Rockstar🎓
Top SchoolPrateek Desai is a Machine Learning Engineer at AWS with over five years of experience building scalable AutoML and benchmarking systems, including major contributions to AutoGluon and the TabRepo project that compares models across 200+ datasets and 1500+ precomputed configurations. He designs high-throughput MLOps infrastructure—creating an EKS cluster for 96-GPU foundation model training, running distributed HPO with Ray, and architecting TabFlow, a SageMaker-based benchmarking pipeline that reduced what would be 600k hours of serial work to ~65 hours. Beyond model training, he specializes in CI/CD, test automation, and cost-optimized infra (cut MLOps spend by 30%), and has improved testing and benchmarking workflows for prominent open-source projects like AutoGluon. A NYU-trained engineer based in New York, he blends systems-level distributed training expertise with practical tooling that makes it straightforward for practitioners to plug in their own tabular models and compare to SOTA.
5 years of coding experience
5 years of employment as a software developer
Bachelor of Technology - BTech, Computer Science, Bachelor of Technology - BTech, Computer Science at College of Engineering PuneMaster's degree, Computer Science, Master's degree, Computer Science at New York University
Github Skills (10)
github-ci10
automated-machine-learning10
githubaction-workflow10
pytest10
python10
cicd10
testing10
bash8
docker6
dockers6
Programming languages (5)
Github contributions (5)
autogluon/autogluon
Aug 2023 - Mar 2025
Fast and Accurate ML in 3 Lines of Code
Role in this project:
DevOps Engineer & QA Engineer / Test Automation Engineer
DevOps Engineer & QA Engineer / Test Automation EngineerContributions:195 reviews, 127 PRs, 63 pushes in 1 year 7 months
Contributions summary:Prateek primarily contributed to improving the testing infrastructure and CI/CD pipeline for the AutoGluon project. Their work included fixing platform tests, adding markers for platform-specific tests, and modifying the CI configuration for different platforms like Windows. They also integrated benchmarking and dashboarding within the CI process. Furthermore, the user addressed test failures and updated test configurations across multiple areas.
forecastingimage-textmlppythonmeta-learning
prateekdesai04/autogluon
Aug 2023 - Sep 2024
AutoGluon: AutoML for Image, Text, Time Series, and Tabular Data
Contributions:104 PRs, 967 pushes, 187 branches in 1 year 1 month
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.