Education - University of Texas, Austin

  • I am a senior, I expect to graduate in May of 2022.
  • BBA, Quantitative Finance
  • BS, Mathematics

Sigmoid Technologies - Co-founder; December 2019 to June 2020

  • I partnered with a friend to design and build an app which allows on the fly diagnosis of chest x-rays.
  • This included work with machine learning which entailed training convolutional neural networks as well as replicating established scientific work such as CheXNet.
  • I also built a backend system on Firebase which includes the ability to authenticate users, upload images, and process data (all done in Python).
  • We also integrated and maintained a NoSQL database, allowing us to efficient serve both our iOS app and our automatic image processing services
  • We faced challenges including getting our ML models to run automatically in the cloud (via Google Cloud AI Platform and Google Cloud Function), and getting these models to run in a performant manner on constrained hardware.
  • In the end, we shut down the project because of business model concerns. It wasn't clear that we could design a device with unit economics that were even close to rerasonable.

Projects and Research

  • Differential Geometry and Finance; February 2020 and onwards

    • This research is on going and as of yet unpublished
    • With the occasional assistance of several professors, I have been investigating the application of differential geometry (in particular, Riemannian geometry) to the study of financial risk
    • This approach yielded a new way of thinking about risk measures, and promises to yield interesting results concerning measurement of arbitrage opportunities
    • Any research program is at times fraught with challenges, and for me that has mostly been in digesting a wealth of literature across fields as far apart as statistical divergences and mathematical physics

  • Triangular Arbitrage in Cryptocurrency Markets; March 2019 to August 2019

    • Worked to find mean reversion in AxB - C, where A, B, C are currency pairs
    • Learned to work with large amounts of data: in aggregate I collected over 7 GB of financial time series data
    • Learned to work with RESTful APIs to communicate with exchanges
    • Discovered a closure on the set of approximations for any given currency

  • Rust vs C++: Prime Factorization; Last Edited Dec 29, 2019

    • Gauged code performance using prime factorization, a process important in cryptography
    • Determined C++ to be slightly faster than Rust, using the same algorithm
    • Wrote a blog post (which can be found here) detailing the methodology and code I used in my tests

Development Skills

  • Python

    • Data manipulation
    • Data gathering via REST APIs and Websocket streams (done on Huobi)
    • Medical diagnostics using machine learning (convolutional nets and dense nets) via Tensorflow and public datasets
    • Familiarity with pandas, numpy, and tensorflow

  • Rust

  • TypeScript

    • I have some basic experiance with typescript, enough to appreciate it as an upgrade over javascript.

  • Development Tools

    • I have experience with Git across GitHub, GitLab, and Bitbucket (this website is hosted on GitHub, actually)
    • I'm also familiar with Anaconda and environment management, on both Windows, Mac, and Linux

  • OS

    • Development on Linux (Ubuntu), Mac, and Windows
    • This includes development on memory and compute constrained platforms, such as Nvidia's Jetson Nano, Google Cloud AI Platform, and Google Cloud Functions

  • Further Skills

    • I have varying amounts of exposure to R, Matlab, SQL, and Bash
    • I've also built computers before, and have some degree of familiarity with computer architecture and performance maximization

Organizations

  • Texas USIT

    • General Member, August 2018 to Jan 2020
    • Participated in weekly Analyst Group sessions, August 2018 to May 2019
    • Participated in 2019 Analyst Group Stock Pitch