Harsh Patel
Software Engineer - AWS AI at Amazon
United States
Summary
👤
Senior🎓
Top SchoolHarsh Patel is a software engineer with 10 years of experience specializing in large-scale ML infrastructure and model training on AWS. At Amazon's AWS AI team he engineered data ingestion and PyTorch integrations that improved training throughput and enabled CUDA 11 support, and developed a proprietary ensemble method to accelerate fine-tuning of T5-3B. He has hands-on experience optimizing IO across FSx, S3, EBS, and EFS, building dataloader plugins, benchmarking multi-node training, and packaging vision models for SageMaker. Past internships at VMware and Teradata show a pattern of delivering GPU-enabled cloud integrations and performance gains, while early work in labeling automation demonstrates practical NLP and serverless skills. Comfortable at the intersection of applied ML and systems engineering, he combines deep technical detail with a focus on measurable production improvements.
10 years of coding experience
3 years of employment as a software developer
University of California San Diego
Github Skills (35)
portable10
pytorch10
docker10
dep10
mxnet10
sagemaker10
aws10
mutation10
dataflow9
javascript9
scheduler9
python9
scala9
machine-learning9
tensorflow9
Programming languages (5)
Github contributions (5)
harshp8l/harshp8l.github.io
Dec 2015 - May 2018
Contributions:12 pushes, 1 branch in 2 years 5 months
harshp8l/mxnet-infrastructure
Oct 2018 - Dec 2018
Temporary repository for MXNet infrastructure code
Contributions:6 PRs, 49 pushes, 6 branches in 2 months
infrastructuremxnet
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.