2026 — Now
I design and own production bioinformatics and execution systems that scale across multiple business verticals. My work focuses on replacing ad hoc coordination and fragile workflows with durable, observable infrastructure that reduces operational risk, enables self-service, and materially improves cost efficiency.
I operate across the stack—from pipelines and execution environments to APIs, databases, and cloud infrastructure—taking ownership of ambiguous, high-risk areas where unclear ownership or short-term fixes would otherwise accumulate technical debt.
Key contributions:
* Built WGS genome alignment pipeline in Nextflow using Nvidia Parabricks GPU-enabled parallelization to enable running high depth (60-120x for tumors) samples for non-Lymphoma cancer types, reliably generating 1TB+ bams.
* Architected event-driven architecture adopted across Custom, Translational Research, and Lymphoma, improving reliability, observability, and recovery while reducing coordination overhead.
* Designed and implemented database-backed workflow persistence, enabling automation and avoiding data-wrangling for bioinformaticians.
* Owned workflow orchestration for Custom, expanding to automate WGS alignment, panel ordering, and validation—minimizing manual hand-offs between engineering, IT, and lab teams.
* Built self-service, human-in-the-loop automation for capture panel ordering and quoting via Slack webhooks, Python/FastAPI, and Twist's API, enabling non-technical stakeholders to complete orders independently.
* Led major pipeline performance and cost optimizations, delivering ~65% per-run cost reductions and cutting runtime by >50% on our variant caller pipeline.
* Identified and eliminated systemic failure modes in data ingestion and logging, preventing silent data loss and recurring cloud cost spikes.
* Serve as a go-to resource for DevOps and infrastructure decisions, contributing to significant cloud cost savings ($43k in Dec 2025 alone) and improved COGS visibility.
2022 — 2023
Spain > Egypt > Ethiopia > South Korea > Canada > Chile
Lived for 8 months in Chile, volunteered at an animal shelter (Centro de Rescate Canino de Ñuñoa) and simultaneously built app prototype to help community members better care for stray dogs in their neighborhoods
Lived for 3 months in Spain; built on proficiency in Spanish
Visited destinations at the top of bucket list including Egypt, Ethiopia, Patagonia
Favorite books read: Atomic Habits, Survival of the Sickest, Hagakure, Adrift, The Corner, Thus Spake Zarathustra, Pandora’s Star and Meditations
Remote - New York/Florida
Quickly learned GoLang/GORM to develop the persistence layer for the lab’s automated work orchestration API. This system resulted in substantial cost reductions for liquid biopsies, as well as shorter Turnaround Time and enhanced ability to forecast COGS.
Utilized Node.js to handle various tasks within the Patient and Provider Portal, including the creation of an Address Validation microservice that served the commercial platform.
Met directly with users and process engineers to conceive User Interfaces for the Lab Systems and Automations division.
Executed frontend dev on UI for Lab/Automation side as well as Patient/Provider Portals.
San Francisco Bay Area - Remote starting December 2019
Defined software requirements through direct interactions with cross-functional teams, including biologists, process & automation engineers, as well as lab technicians.
Implemented Flask API for automated production planning and chip filling logic (used by Production Planning UI written by Kipp McMichael), maximizing use of disposable materials like silicon chips & inks that are scrapped after each run, decreasing 8+ hour manual planning sessions to around 10 minutes.
Developed separate synthesis execution service in C#/.Net Core to pre-process synthesis run data & break up the Synthesis Execution+Production Planning monolith.
Optimized database usage, with Redis for ephemeral data and AWS S3 for larger entities, dropping two tables that slowed down system-wide queries & consolidating data models.
Centralized data access and integrated the QC pipeline with MES for ISO compliance.
Improved TAT with error handling in a previously error-prone production chokepoint.
Streamlined QA by enabling the creation of test data from real orders, eliminating the overreliance on costly/disruptive "wet runs" using production instruments and materials.
Supported the integration of new, complex products, including multi-part genes (<=5Kilobase) and large NGS panel orders that required more than one synthesis run.
Created UI with high-res image viewer for job run tracking during COVID-19 restrictions.
Developed a C# Materials Storage microservice for real-time tracking of items in the 24/7 production line, reducing ad-hoc decisions in a lab environment and optimizing utilization.
Designed & implemented Materials Storage UI, using barcode scan/DeviceOrientation events to avoid keyboard use in the lab, with SignalR/socket.io for real-time updates.
Implemented software workarounds to synthesizers’ (hardware-based) biases/failures.