I turn ambiguous problems into working software.

What I actually do.

I'm a software engineer working across backend systems, applied ML, and cloud infrastructure. Currently building PHI validation and data pipeline tooling at Asembia, after building an AWS IAM verification framework at Stevens and shipping AI SaaS backends at Headstarter.

Years of experience

2+

AWS misconfigs reduced

40%

IAM roles verified

300+

Data Analyst · Asembia LLC

Mar 2026 – Present

• Engineered a Python-based PHI validation system replacing a fully manual QC process across 30+ specialty pharmacy data feeds, automatically verifying field-level hashing, randomization, and ID masking against a database-driven configuration, cutting validation cycle time by 30%
• Built end-to-end Rhapsody integration workflows automating file pickup, data transformation, and database ingestion for specialty pharmacy feeds, saving 2–3 hours per analyst per cycle
• Architected and maintained SQL Server stored procedures powering multi-pass QC validation across pharmaceutical manufacturer reporting pipelines, surfacing 15+ data integrity issues per cycle that previously reached downstream systems undetected

Software Engineer · Stevens Institute of Technology, Department of Computer Science

Oct 2024 – Mar 2026

• Reduced AWS cloud misconfigurations by 40% across 300+ IAM roles, KMS policies, and VPC security groups by building a Python verification framework using Terraform and AWS SDKs, with findings reviewed in weekly cross-team code reviews
• Improved cloud access control detection accuracy to 83.8% and cut audit time by 70% by engineering automated evaluation and reporting workflows, delivered iteratively across agile sprint cycles
• Co-authored paper accepted to NLBSE'25 on IAM policy synthesis via LLMs by building evaluation pipelines and analysis tooling in Python

Software Engineer Fellow · Headstarter AI

Jul 2024 – Sep 2024

• Achieved consistent deployment reliability across 3 production AI SaaS applications by building containerized backend services with Docker, GitHub Actions CI/CD, and AWS/GCP infrastructure
• Increased search query precision by 30% over a 100K+ review corpus by engineering NLP-based feature extraction, relevance ranking, and structured data pipelines