MH Naderi

MSc in Computer Science

About Me

Hi! This is Mohammadhossein (MH), a software engineer in Toronto working at the intersection of distributed systems and machine learning.

I build systems that make training and running machine learning models faster and more reliable at scale. Most recently that was AdapTrain, the focus of my MSc research at York University’s PACS Lab under Professor Hamzeh Khazaei. It is a system that adaptively partitions neural network training across heterogeneous compute nodes based on real-time performance signals, and it improved training efficiency by up to 8x over baseline Independent Subnet Training.

I built that work end to end: the partitioning and scheduling logic, a model that predicts training time from network configuration, the GPU and system-level instrumentation feeding those decisions, and the evaluation pipeline that validated the results, all deployed on Kubernetes.

Before graduate school I earned my BSc in Computer Engineering from Amirkabir University of Technology (Tehran Polytechnic), specializing in software engineering and artificial intelligence under Professor Hamid Reza Zarandi. I also spent time as a full-stack developer at Fanap Telecom, building REST APIs and Spring Boot microservices for an enterprise OSS/BSS platform, and as a research intern at the HPC Lab at IPM, working on parallel programming with C++ and CUDA.

Right now I am building a distributed agent orchestrator, which extends that same interest in adaptive distributed systems into agentic AI: a LangGraph orchestrator routing tasks across specialized MCP agents, each running as its own Kubernetes service with Prometheus and Grafana observability across the whole system.

I am currently open to Software Engineer, Machine Learning Engineer, and AI Engineer roles.


Areas of Expertise

  • Distributed Machine Learning: model partitioning, heterogeneous compute, adaptive training

  • ML Systems Optimization: performance modeling, scheduling, scalability, fault tolerance

  • Cloud & Cluster Computing: Kubernetes, Helm, container orchestration, resource management

  • Observability & Performance Monitoring: GPU and system-level instrumentation, Prometheus, Grafana

  • Software Engineering: REST APIs, microservices, full-stack applications, testing and CI/CD


Technical Skills

  • Languages: Python, Java, JavaScript/TypeScript, C, C++, Kotlin, SQL, R

  • ML & Data: PyTorch, scikit-learn, NumPy, pandas, SciPy, XGBoost

  • Infrastructure: Docker, Kubernetes, Helm, Jenkins, CI/CD, Git, Linux

  • Backend & Web: Spring Boot, Node.js, Express, Flask, Angular

  • Databases: PostgreSQL, MySQL, MongoDB

  • AI-Assisted Development: Claude Code, Cursor, GitHub Copilot


Explore My Work

  • Work Experience: Professional and internship history

  • Research: Research experience and interests

  • Projects: Selected academic and personal projects

  • Teaching: Courses and teaching experiences

  • CV: Full academic résumé