I build software for projects that I think can do good in the world, working as a tech lead and engineer. I like to work full stack, working on architecting complete products rather than focusing on a single aspect. Product comes first and UI/UX cannot be ignored as an afterthought. Don't undervalue your designers, product managers, and user experience people.

I write code in many languages (Python, Typescript/Javascript, Go, Java, Racket, C/C++, Objective C, Swift, ...) and learn as needed, valuing general principles and design over deep knowledge of a single language. Appreciator of functional programming / style when possible. Simplicity, readability and ability to communicate your design to others is more important than focusing on the cleanest or most efficient code in most cases.

Currently I work by day as a Software Engineer at Uniswap Labs.

Previously I worked at Snapchat on projects including:

• Being a tech lead for the Minis platform, focusing on new verticals, Private Components, and first party Minis.
• Powering Snapchat's civic features to build features such as Register To Vote, Run For Office, and Before You Vote
• Creating unique physical activations for Snapchat’s marketing and events team including museum exhibits, sticker and photo printers, vending machines, and body tracking AR activations using a wide range of technologies.

A collection of various technical projects from my college days, some available on my Github:

• racket-sourcery - A collection of libraries (web-sourcery, sql-sourcery, json-sourcery) aimed at making quick backend web development in Racket easier. Put on indefinite hold, but I always love functional programming and think this type of library is strongly needed. Modeled loosely off python's combination of Flask/SQLAlchemy/json
• Cooperate - a hackaton project focused on reviewing internships and co-op's for college students. Incomplete but I still believe in the idea and may come back to it if I have time.
• A cryptocurrency trading bot capitalizing on basic AI (linear regression + genetic algorithms) and volatility. Closed source as I don't morally think it should be used.
• gpu-no-sql - a toy project to test out GPU acceleration on a NoSQL data model. Works surprisingly well over shallow trees, probably long broken since last run.

These projects are no longer in development, but if you have a passion for any of these I more than welcome someone else to take one over. Please reach out if you're interested!

All opinions expressed on my website are my own and not my current or past employers :)