Shrey Joshi

I'm a software engineer interested in systems programming, machine learning infrastructure, and the places where the two meet. I recently finished a B.A. in Philosophy with a Computer Science minor at the University of Texas at Dallas, where I studied on a full-ride National Merit scholarship.

My research background is in applied machine learning for geoscience. Working with Dr. Ellen Rathje at UT Austin, I built GLAS, a global landslide analytics system trained on GLIF — a dataset compiled from roughly 200GB of geophysical data across six sources. That work took a first grand prize at ISEF and was published at the 2021 IEEE MIT URTC.

Since then I've worked as a software engineer at PwC and Minion AI, and I spend most of my free time writing Rust: a chess engine built on magic bitboards, a terminal video-calling client that renders frames as sixel graphics, and the FFI bindings underneath it.


Education
  • University of Texas at Dallas
    University of Texas at Dallas
    B.A. Philosophy, minor in Computer Science
    Full-ride National Merit Scholar
    Aug. 2022 - May 2026
Experience
  • PwC
    PwC
    Software Engineering Intern
    Jun. 2024 - Aug. 2024
  • Minion AI
    Minion AI
    Software Engineer
    Jan. 2023 - Feb. 2023
  • University of Texas at Austin
    University of Texas at Austin
    Machine Learning Researcher
    Aug. 2020 - Apr. 2022
Honors & Awards
  • Codon Digest Winner, Bio x ML Hackathon
    2023
  • National Merit Scholar (full ride)
    2022
  • ISEF 1st Grand Prize ($10,000)
    2021
Selected Publications (view all )
GLAS: A Global Landslide Analytics System
GLAS: A Global Landslide Analytics System

Shrey Joshi*, Ishaan Javali*, Ellen Rathje (* equal contribution)

IEEE MIT Undergraduate Research Technology Conference (URTC) 2021

A scalable, low-latency system for global landslide forecasting, severity assessment, and date estimation, built on GLIF — the first publicly available dataset of Global Landslide Incidents and Features, compiled from roughly 200GB of elevation, climate, lithology, forest-change, and infrastructure data across six sources.

GLAS: A Global Landslide Analytics System

Shrey Joshi*, Ishaan Javali*, Ellen Rathje (* equal contribution)

IEEE MIT Undergraduate Research Technology Conference (URTC) 2021

A scalable, low-latency system for global landslide forecasting, severity assessment, and date estimation, built on GLIF — the first publicly available dataset of Global Landslide Incidents and Features, compiled from roughly 200GB of elevation, climate, lithology, forest-change, and infrastructure data across six sources.

All publications