Résumé

Felix Hirwa Nshuti

I am a Felix Hirwa Nshuti. I work at the intersection of computer architecture, compilers, and machine learning, with a focus on building efficient systems for AI workloads. I explore how compiler optimizations, hardware-aware scheduling, and specialized architectures can accelerate deep learning model training and inference. I am an open-source contributor and maintainer who likes the community interaction.

I am currently pursuing MS in Electrical and Computer Engineering (ECE) at Carnegie Mellon University (CMU).

Previously, I completed a Bachelor's degree in Computer Science and Engineering from Pandit Deendayal Energy University (PDEU). I concentrated on the deep understanding of the computer architecture, Programming Languages and Compilers, and did various internships in Machine Learning field.

In my free time, I play football and share memes with friends 😎

profile photo

Projects

  • TransISA

    A lightweight transpiler for translating assembly code from x86-64 to RISC-V ISA. Using traditional compiler pipeline techniques, it performs lexical analysis, parsing, semantic analysis, and code generation.

    This tool is mainly designed to bridge the gap between x86 legacy code and RISC-V ISA. It is a collaborative work carried out to satisfy major project requirements of the eighth semester at PDEU.

  • GSoC2024 project

    At sktime, A unified framework for machine learning with time series

    Implemented adapter for darts regression models and pytorch classification models GRU and GRU-FCN.

  • Struco

    Authors: Felix Hirwa Nshuti & Mellisa A. Gblinwon

    A simple tool designed to extract the Intermediate Representation (IR) and Control Flow Graph (CFG) form C/C++ and Python source code. This is collaborative work carried out to satisfy minor project reuirements of the seventh semester at PDEU.

Some Open-Source Contributions

  • Ivy - A Unified AI Framework (Maintainer: 2022 - 2023)
  • pytorch-forecasting - Time series forecasting with PyTorch (Maintainer: 2024 - Present)
  • sktime - A unified framework for machine learning with time series. (Maintainer: 2024 - Present)
  • pytorch-lightning - Pretrain, finetune and deploy AI models on multiple GPUs, TPUs with zero code changes. (Contributor: 2024 - Present)