Ran around and did lots of things in the last four years here --- Implemented automatic model partitioning for multi-gpu inference; built this on top of a system created by our R&D team creating a Virtual Tensor abstraction that streamed PyTorch ops to remote workers. (Python)- Closely interfaced with open-source PyTorch 2.0 developments, filed issues, and contributed fixes. (Python)- Handled, ran, profiled, and debugged customer models and open-source models directly, in various frameworks including PyTorch, ONNXRuntime, TensorFlow, and TVM.- Drove model deployment and accuracy validation initiatives, both internally and for external hardware companies. (Python)- Significantly reduced model optimization strategy search time via design of a tuning records caching service. (Rust, Python)- Bootstrapped visualizations and frontend interactions, from scratch. (React, JS, vega)- Helped bootstrap the SaaS platform, from scratch. Defined database schemas, created APIs, defined protobufs, worked on all components of the whole e2e flow: client -> api server -> scheduler -> worker. (Rust)- Did a lot of random terraform.- Iterated through alternative prototypes to discover product-market fit.- Hacked on the TVM compiler -- improved coverage and added metadata passes. (C++)- Designed and created a weekly integration model testing service. (Python, Kubernetes)

Contribute to algorithm design and optimization research in the Programming Languages research group at the Paul G. Allen school. Specific work regards program synthesis. contributed to paper accepted to ASPLOS '19: https://homes.cs.washington.edu/~emina/doc/swizzle.asplos19.pdf

Aid students in achieving a better understanding of Hardware-Software Interface material: C, caching, memory, x86-64 assembly.

Added optimization pass to the open-source Android bytecode re-optimizer redex: detected and removed unused arguments, reducing the uncompressed Facebook app for Android by 130KB. https://github.com/facebook/redex/tree/master/opt/remove_unused_args(C++)Designed and implemented basic infrastructure for exposure of optimization metadata in internal Facebook code file browser view. (React js, PHP/Hack, Python, C++, MySQL)

Revamped and increased robustness of internal graphical inventory location mapping test tool with integration into a generalized testing framework. (C# MVVM, xaml)Improved and maintained a client-available testing tool, with filtering options. (JavaFX)

Verified brain imaging parameters and visualized data in R using the Shiny package.