Co-developed generative AI COATI-LDM research

Built hybrid ML platform using on-prem cluster and cloud resources using Redis, Slurm, Docker, Enroot, Prefect

Implemented molecule conformer generation and energy calculation workflow using AWS Batch - pipeline was used to generate data for COATI paper

Deployed docking pipeline using various ML and FF based methods

Deployed ML platform using Slurm, XGBoost, and on-prem compute

Using Pandas, Docker, and Amazon Sagemaker, S3, and EC2 as part of data science pipeline with a primarily Python codebase

Implemented caching solutions to improve performance of data science dashboard

Built clinical data visualization tools

Leading efforts to refactor existing codebase to adhere to best practices and to create a robust and sustainable code architecture

Parallelized data loading processes to greatly increase pipeline performance

Created test suite to allow for smoother and more robust deploy process

Worked with bioinformatics scientists to bootstrap and guide development of new Bioinformatics dashboard

Upgraded parts of existing Bioinformatics pipeline to integrate with new order platform

Python, Typescript, AWS, SQL