I'm a software engineer with a background in full-stack development and software engineering research.
Sometimes that means building a CV platform with its own editor and document engine. building a CV platform with its own editor and document engine designing persistent memory and knowledge graph for coding agents. retrieving 289,000+ manga for oddly specific recommendations. mining 482 repositories for a research paper.
Download Resume See My ProjectsSelected work
Products, tools, and experiments built around making complicated things easier to use.
Aergia
Project demo
A CV builder that makes the path from editing to a finished PDF feel direct.
Project Tracker
Project demo
A reusable memory layer that helps coding agents keep project context close to the work.
MBuddy
Project demo
A unified media database for unusually specific recommendations.
Career
A path from building client products to researching how software can be made better.
Dalhousie University · Halifax I worked at the intersection of software engineering and simulation research, focusing on code quality, refactoring practices, and practical improvements to research software.
Brain Station 23 I built and maintained client-facing web and mobile products while working across application development, deployment, and legacy codebases.
Overleaf's new compile-time limits make working with slightly large papers more hassle than worth. This walkthrough of the VSCode + TeX Live setup, including some troubleshooting tips will make switching from overleaf easier
What the append-only tool feels like to use day to day: the new loop, the OpenCode plugin's three hooks, the walkable audit trail, and the sharp edges that still need attention.
We had been working around four problems at once with the single-writer assumption. The current tool drops the assumption, the counter, the in-place mutation, and the committed graph in a single architectural shift.
How a project moved onto the tracker, what daily life with it looked like, and how agents picked it up on their own.
Academic work
I study how software quality and refactoring shape the maintainability of simulation software, with additional work in applied machine learning.
Master’s thesis examining software quality problems, code-smell patterns, and refactoring effectiveness across 482 open-source repositories.
Study of how effective and risky refactoring is across simulation modelling systems.
Empirical analysis of code smells across 155 simulation and 327 traditional GitHub projects.
Comparison of classic machine-learning methods and a neural network for personalized city recommendations.
Neural-network approach to classifying Alzheimer’s disease from brain MRI data.
Background
Dalhousie University
BRAC University
Have a question, collaboration idea, or just want to say hi? Reach out through any of the platforms below.
© 2026 Riasat Mahbub. All rights reserved.