About

I'm a senior at Grinnell College studying computer science with a concentration in statistics (graduating May 2027), and I'm looking for a new-grad software or data engineering role in health tech, medical devices, or healthcare more broadly.

My mom has lived with An autoimmune condition where the body stops making insulin entirely. Not caused by diet or lifestyle, and not the same disease as type 2.Click to read more for most of her life, and growing up I watched how directly her health depended on a stack of devices actually working. The insulin pump, whose algorithm re-evaluates every five minutes and adjusts her basal rate and delivers correction boluses to hold a target glucose, and which, depending on the generation of the system, either kept her steady over the long term or didn't. The continuous glucose monitor, whose accuracy still varies meaningfully between brands, and within a single sensor between its first day and its last. The new infusion set every few days, the new sensor about every week, and the two-hour warm-up before a fresh one will report anything. The pairing between devices that may or may not work out of the box. Even ordering supplies, through a website that adds friction to something she has no choice but to do.

Seeing all of that up close, and watching which parts improved over the years and which stagnated, gave me a close-up view of what still needs to be built in healthcare technology. It's the reason I built refillr, an iOS app for tracking and refilling medications, and the reason I want to keep working on this: better interfaces, better data infrastructure, tools that ease the burden of managing a chronic condition around the clock.

This past summer I was a data engineering intern at Eli Lilly and Company, on the Consumer Data Engineering team within Business Insights & Analytics, working on how a global team builds A dataset built and maintained like a software product: owned, documented, versioned, and reliable enough for other teams to build on. with AI systems that plan and carry out multi-step tasks on their own, calling tools and making decisions along the way, rather than answering a single prompt. in a regulated environment.

Outside of classes, I'm on the leadership team of Girls Who Code at Grinnell; we run a weekly after-school club at the local middle school covering programming fundamentals and robotics, and mostly getting girls excited about computer science.

I'm available starting June 2027.

Experience

Eli Lilly and Company

Summer 2026

Data Engineering Intern · Consumer Data Engineering, Business Insights & Analytics

  • Defined AI-assisted development standards for BI&A's agentic data product initiative: standards that hold up in a regulated, Protected Health Information: any health data that can be tied to a person. Regulated under HIPAA in the US, so systems that touch it have strict rules about access, handling, and audit.-sensitive environment, with Writing a precise, reviewable specification of what a system should do before any code is written, then building (and having AI build) against that spec rather than against a conversation. at the core.
  • Evaluated spec-driven development frameworks (Superpowers, GitHub Spec Kit, and the An open-source framework for AI-assisted software development. It gives AI agents defined roles (analyst, architect, developer, and so on) and structured workflows so they build from specs and stories rather than free-form prompts. Docs →) for enterprise fit and for fit with the team's existing day-to-day workflows, and made the recommendation the team adopted.
  • Architected and built a BMad module for the team's A cloud platform for storing and processing large datasets, built around Apache Spark. Widely used for enterprise data engineering. data product pipeline, covering the full cycle: vendor intake, Comparing a new data source against the existing table structure to see what fits as-is and what needs a new column or table., spec generation, story creation, review, and User Acceptance Testing: the final check where the people who requested the work confirm it does what they asked, before it goes to production. sign-off. It replaced a manually maintained, non-version-controlled Excel source-to-target mapping with a human- and machine-readable spec that lives in Git, can be updated by AI, and holds transformation logic out by design, so the spec stays a stable interface between requirements and the downstream Extract, Transform, Load: the code that pulls data from a source, reshapes it, and writes it into the destination tables. implementation.
  • Estimated by the team's leads to cut data product delivery time by two to four weeks, by catching mapping discrepancies at spec review rather than at UAT, where a catch sends the modeling team back through a full rework cycle.
  • Built with BMad's own module-builder tooling; packaged, validated, and wrote cross-platform install scripts for macOS, Linux, and Windows. Delivered on a review branch as a working Minimum Viable Product: a first working version with the core functionality in place, built to be tested and extended rather than to be final. for the team to carry forward.
  • Wrote and produced a narrated explainer for the module with Remotion and ElevenLabs, storyboarded across nine scenes with an automated layout audit so the design reference couldn't drift from the video.
  • Worked on a globally distributed agile team of 20+ engineers, tracked in Jira, across roughly 11 vendor sources covering prescription claims, call records, and medication dispense history. Day-to-day in GitHub and Claude Code, with conventional commits and PR review throughout.

Projects

refillr

iOS · Swift · in development

An iOS app for tracking, managing, and refilling medications and supplements, with a customizable interface backed by the NIH's dietary supplement label and prescription drug databases.

View on GitHub →

Non-Human Vision

Unity · C#

A VR experience for Meta Quest 3 that simulates how animals and insects perceive their surroundings, inspired by the toBeeView research project and built by a five-person agile team in Software Design & Development.

View on GitHub →