System Optimization: Revolutionized data pipeline efficiency via Redis caching, achieving a 10X reduction in Kubernetes resource consumption.

Developer Productivity: Built an automated CI/CD pipeline using Python, Bazel, and Tilt, reducing deployment snapshot creation time by 90%.

Containerized scripts using Docker and deployed AWS ECS tasks on AWS Fargate to perform monitoring.

Collaborated with team members to redesign middleware messaging and set up fanout with AWS SNS and SQS.

Dived deep into credential and connection configurations of AWS services, PostgreSQL, and Docker containers.

Reported vulnerability counts for 1M+ results of port scanning for evaluation and analysis of cyber risk.

Preprocessed 20K+ retail store information using PySpark and SQL, improved script performance by 40 folds.

Extracted and engineered features through Natural Language Processing to generate similarity scores.

Built Random Forest Classifier for the recommendation, recursively trained a labeling model starting with 80 search queries to label 5K+ pairings, reached AUROC of 95%, precision of 82%, and recall of 73% on test data.

Led 3 students to build Regression models with Python and R to identify short tandem repeats (STRs) impact with gene splicing events, targeted 2K+ highly confident associations significant to rare diseases studies.

Engineered gene splicing features from 230K+ locations and 18K+ samples and associated splicing metrics with STRs to identify 22K+ statistically significant splice-STR pairs out of billions of genomic data entries.

Optimized the pipeline workflow by splitting datasets to 1K+ subsets and parallelized the tasks using array jobs on High Performance Computing clusters, improving the total performance by 1,200 times.