Where I've worked.

Senior Software Development Engineer in Test Nov 2023 – Present

Leading GenAI quality strategy for SAS Health, an enterprise healthcare data platform.

• Built an agentic engineering system integrating Playwright CLI, GitHub MCP, Jira/Confluence MCP, and a custom Zephyr MCP server, presented during R&D demos to 300+ engineers and earned recognition from the senior leadership team including the CTO and SVP of Engineering, accelerating cross-team communities of practice for sharing and collaborating on agentic engineering practices.
• Architected ETL validation pipelines across 5 cloud ingestion surfaces (Azure FHIR, AWS HealthLake, Azure Blob, AWS S3, FHIR Bundle transforms) and 3 database layers (Snowflake, PostgreSQL, SingleStore), catching 30–40% of data-contract regressions before production.
• Designed a modular automation framework from scratch with embedded WCAG 2.1 audits and OMOP/FHIR/CDISC validators in CI/CD, standardizing testing across 3 product teams and cutting duplicated test infrastructure effort by ~40%.
• Built a multi-layer E2E suite for SAS Health Copilot, the natural-language clinical data discovery agent that lets clinicians and researchers explore EHR/FHIR data and identify patient cohorts without SQL, covering eval, API, and UI layers across the multi-agent pipeline.
• Validated RAW-to-SDTM clinical data transforms for SAS Life Sciences, the CDISC-compliant pipeline that converts raw trial data into FDA/EMA submission-ready SDTM datasets, reaching ~95% scenario coverage across multiple SDTM domain types and versions.
• Mentored 3 junior SDETs on FHIR/OMOP/CDISC standards, cutting average domain onboarding from ~6 weeks to ~3 weeks.

Software Development Engineer in Test II Nov 2022 – Jun 2023

Built test infrastructure for Nuna’s 2.0 platform launch, a healthcare data platform processing real-world evidence on Protected Health Information.

• Cut smoke-test runtime by 30% via storageState auth caching, parallel test splitting, and concurrent suite execution, directly accelerating feature velocity for the 2.0 launch.
• Built a custom Playwright-to-Slack reporter surfacing GitLab Job IDs, HTML report links, and SDET-tagged failures in real time, eliminating ~2–3 hrs/week of manual triage across the team.