About

Hello! I am Mohammad Moridi, a Software Engineer at Microsoft in the AI Frameworks team, based in Vancouver, Canada. I specialize in cloud-based fault-tolerant systems, with expertise in C++ programming, multi-core environments, and working with CPU, GPU, and NPU architectures. In my role, I focus on designing and developing AI software in C/C++ and Python, collaborating across hardware and ML teams to optimize AI models for large-scale training and inference on Microsoft AI accelerators.

I hold a MASc degree in Computer Software from the University of Waterloo, where I specialized in persistent memory platforms. Previously, at Huawei, I led database engine optimization initiatives and developed NPU-level operators for LoRA serving on vLLM. I am passionate about contributing to AI innovation and high standards of engineering within an inclusive, collaborative culture.

Basic Information
Email:
moridi@uwaterloo.ca
Location:
Vancouver, British Columbia, Canada
Experience
Microsoft Logo
Software Engineer II

Nov 2024 - Present | Microsoft | Vancouver, Canada

AI Frameworks: Developing and optimizing software for large-scale AI model training and inference on the Microsoft AI accelerators using C/C++ and Python.

Collaboration: Working closely with cross-functional teams, including hardware and ML, to create efficient AI solutions across diverse platforms.

Huawei Logo
Senior Research Engineer B

Sep 2023 - Nov 2024 | Huawei Technologies Canada Co., Ltd. | Vancouver, Canada

AI/ML: Designed and implemented customized kernels for Ascend-NPU to optimize LLM inference on vLLM.

Database: Contributed to research initiatives and early-stage engineering of an in-memory OLAP engine for cloud services.

University of Waterloo Logo
Graduate Research Assistant

Sep 2021 - Sep 2023 | University of Waterloo | Waterloo, Canada

Research experience on shared memory algorithms.

Designed and implemented snapshotting mechanisms for persistent memory-mapped files and detectable objects for persistent memory.

Graduate Teaching Assistant

Jan 2022 - Apr 2023 | University of Waterloo | Waterloo, Canada

Helped in holding courses in Algorithm Design, Database Systems, Systems Programming, and Data Abstraction.

Huawei Logo
Blockchain Researcher Intern

May 2022 - Aug 2022 | Huawei Technologies Canada Co., Ltd. | Waterloo, Canada

Worked on a collaborative project called Towards a High-velocity Permissioned Blockchain.

University of Tehran Logo
Undergraduate Teaching Assistant

Oct 2018 - Jul 2021 | University of Tehran | Tehran, Iran

Assisted in more than 15 courses such as Advanced Programming and Operating Systems.

Undergraduate Research Assistant

Oct 2019 - Feb 2021 | University of Tehran Science & Technology Park | Tehran, Iran

Analyzed social network behavior in the Ethereum network using graph analysis.

Machine Learning Intern

Jul 2019 - Sep 2019 | University of Tehran Science & Technology Park | Tehran, Iran

Participated in several Kaggle competitions and developed solutions using Scikit-Learn.

Top Skills
C/C++
Python
Low-Level Programming
Multi-Core Programming
Linux
Java
Education
University of Waterloo Logo
Master's degree, Computer Software Engineering

Sep 2021 - Sep 2023 | University of Waterloo | Waterloo, Canada

Grade: 94 / 100. Thesis on snapshotting mechanisms for persistent memory-mapped files. View Thesis

University of Tehran Logo
Bachelor's degree, Computer Engineering

Sep 2016 - Jul 2021 | University of Tehran | Tehran, Iran

Grade: 90 / 100. Thesis on social network analysis within the Ethereum ecosystem.

Publications
Snapshotting Mechanisms for Persistent Memory-Mapped Files

Published in ApPLIED@PODC 2024. DOI: 10.1145/3663338.3665832

Investigates ways to enhance the reliability of persistent memory systems, focusing on snapshotting mechanisms and their role in system resilience. Introduces new snapshotting consistency models and mechanisms to improve performance and enhance system responsiveness. Provides experimental analysis demonstrating throughput and latency improvements.

A Closer Look at Detectable Objects for Persistent Memory

Published in ApPLIED@PODC 2022. DOI: 10.1145/3524053.3542749

Explores the adaptation of multi-core algorithms to persistent memory, introducing the "Unified Detectable Sequential Specification" (UDSS), which simplifies interfaces and coding. Experiments conducted using Intel Optane memory demonstrate the performance implications of the implementation.

Certificates
University of Waterloo Logo
ExpecTAtions Teaching Assistant Workshop

Issued by University of Waterloo in Oct 2021

IEEE University of Tehran Student Branch Logo
IEEE Data Science Winter School as Machine Learning Instructor

Issued by IEEE University of Tehran Student Branch in Jan 2020

IEEE University of Tehran Student Branch Logo
IEEE Data Science Winter School as Statistical Inference Instructor

Issued by IEEE University of Tehran Student Branch in Jan 2020

University of Tehran Logo
Workshop on Teaching Assistant Training

Issued by University of Tehran in Apr 2019

Honors and Awards
Huawei Logo
2024 Q2 President's Spot Award

Issued by Huawei Technologies Canada in Oct 2024.

Was awarded the Best Team Award for outstanding contributions to LLM serving performance, by the President of the Canadian Research Center.

ETHGlobal Logo
ETHGlobal Waterloo 2023 Finalist

Issued by ETHGlobal in Jun 2023.

Project: Smarter Contract. View Project

ETHGlobal Logo
Winner of Best Usecase of Hyperlane

Issued by Hyperlane in Jun 2023.

Project: Smarter Contract.

Waterloo Blockchain Logo
First Place in the Axelar Track at OlympiHacks

Issued by Waterloo Blockchain in May 2023.

Project: Glue. View Project

University of Waterloo Logo
International Master's Award of Excellence

Issued by University of Waterloo in Sep 2021.

View Description

University of Waterloo Logo
Graduate Research Studentship

Issued by University of Waterloo in Sep 2021.

View Description

References
Contact Me

Location

Vancouver, British Columbia, Canada