-
📫 LinkedIn
-
🚀 Current (since Aug 2021): Staff Software Engineer at Community Tech Alliance.
- Founding engineer with roles spanning people management and technical leadership.
- Skills:
- Software Engineering, Data Engineering, Cloud Infrastructure (GCP and AWS), and full-stack (Django and React).
- Python, SQL, BigQuery, Airflow, dbt, Google Cloud Platform, AWS, Docker, Kubernetes, Terraform, Django, React
- People management, technical leadership, architecture design, team development, mentorship
-
👀 Previous:
Click to expand
- 🏋️♀️ 2021: Staff Data Engineer at MyFitnessPal – Built pipelines to support product analytics.
- 🇺🇸 2020: Data Engineer at Biden for President – Engineering support for Paid Media Analytics and Election Night Reporting.
✈️ 2017-2020: Staff Data Analyst at Expedia – Business performance insights with innovative anomaly detection strategies.- 📊 2015-2017: Senior Data Analyst at MyFitnessPal – Dashboards and experimentation.
- 🧐 2015: User Researcher at MyFitnessPal – User interviews, user testing, surveys, weird queries.
- 🎓 2011-2015: PhD Student at Harvard University (Psychology) – Writing and speaking (but fancy).
These are for fun and education - or at least they better be, because I'm sure not making any money off of them.
🐐🗒️ Notes for Goats
I wanted a personal note-taking tool with some features but not all the features, and certainly not for any dang $10/month. Good grief! So I vibe-coded my own tool. For local deployment ONLY by serious people ONLY.
✂️ List Cutter
A web app that helps you cut lists from CSV files using user-provided filters provided in SQL-like WHERE logic.
Django backend, React frontend, previously deployed on AWS EC2 but then I spent all my free tier credits, so this is mostly a historical relic of a fascinating time in my life.
This is NOT a CRM.
A web app that uses the Chess.com API to do things. Still a work in progress. My vision is to enable users to subscribe to chess.com players and receive email notifications when they are in a live match!
Built with a Django backend and React frontend using Material-UI, containerized with Docker and (previously) deployed on AWS EC2.
This web application uses the power of (so-called) "Artificial Intelligence" (dubious!) to query Reddit and summarize the content of posts in order to "answer" questions. (Note: this really doesn't work well at all, but I had fun playing with the OpenAI and Reddit APIs.)
Stay tuned lmao
"Dead Internet Theory" states that the internet increasingly consists of AIs interacting with each other, leading us towards a technological event horizon in which all content and interactions are no more than a never-ending cycle of recursively generated AI slop. What would that actually be like? Let's find out!
Dead Internet is a game and/or simulation and/or work of performance art in which the user creates AI personas, or rules that generate AI personas, and then unleashes them onto a (fictional!) online message board. For bonus points, this project would make the message board publicly accessible (no longer fictional!), enabling both IRL bots and IRL humans to join in the fun. Neat!
Psychologists sometimes joke about doing "me-search" when their research could be construed as being self-directed or self-serving - for example, a divorced person might study outcomes of divorced people to see what lies in wait for them. Awkward!
Anyway, if you had a whole bunch of personality tests - including links to research about their validity (we report you decide!) - then you could build a profile for yourself that helps you understand your deepest, truest, most arrestingly honest nature of your being. You could evaluate peer-reviewed research before choosing which personality measurements to use, and/or just mess around and find out what your Disney Princess archetype says about your Fursona.