Software engineer & CS studentBased in New York City

Muhammad

I build software.
Then I figure out how to make it better.

Backend systems, full-stack applications, and the details that make them dependable.

SELECTED WORK / 2026Built with curiosity. Refined with tests.

01 / SELECTED WORK

A few things I’ve built.

Real projects. Interesting problems.
A closer look under the hood.

Engineering diagram · not a product screenshot

PROJECT 01Live

WatchMate

A home for your next watch.

A movie and TV tracking platform I independently designed, built, and deployed — from the Spring Boot API to the React interface.

100
users in two months
4,298
episodes tracked
420 → 28 ms
movie-detail latency with Redis
  • Hashed, atomically rotated refresh tokens, JWT authentication, and Bucket4j rate limiting.
  • Large multi-season imports run in the background, with HTTP 202 responses and polling to keep the interface responsive.
  • About 90% line coverage with JUnit and Testcontainers; testing and OCI deployment automated with GitHub Actions and Docker Compose.

In a 10,000-request benchmark: 80% cache hits and 75% fewer TMDB requests.

  • Java 21
  • Spring Boot
  • Spring Security
  • MySQL
  • Redis
  • React 19
  • TypeScript
  • Docker
  • GitHub Actions

Engineering diagram · not a product screenshot

PROJECT 02In development

NYC Housing Intelligence

Good answers start with reliable data.

An autonomous pipeline for publicly available NYC housing data. The first working foundation: discovering, validating, and importing MapPLUTO releases into PostGIS.

856,614
tax-lot records imported
230
passing tests, including 85 integration
0
duplicates on repeat imports
  • Discovers releases and downloads official ZIPs autonomously, verifying integrity, schemas, and geometries before ingestion.
  • Transactional staging-to-silver merges are idempotent: repeat runs create no duplicates or unnecessary updates.
  • Advisory locks, retries, rollback, and recovery protect failed or overlapping imports.

Still ahead: additional datasets, address resolution, and grounded AI functionality. These are planned, not completed features.

  • Python
  • PostgreSQL
  • PostGIS
  • GDAL
  • SQLAlchemy
  • Alembic
  • Docker

Engineering diagram · not a product screenshot

PROJECT 03Academic team project · 3 engineers

Renter Dispute Assistant

Making the next step clearer.

A team-built case-management application for rental disputes, with evidence from both parties and official New York forms prepared for review.

  • Contributed authenticated case workflows, case-level access controls, evidence-image uploads, and landlord responses.
  • Built filing-path selection and prefilled official New York PDF forms for review and download.
  • Integrated structured, nonbinding Gemini assessments with timeouts, image limits, and evidence from both parties.
  • Node.js
  • Express
  • MySQL
  • React
  • Gemini API

02 / EXPERIENCE

Shipping with a team.

March 2025 – May 2025

Software Engineering Intern

DNR Events LLC

Worked in a four-engineer Agile team on DNRcast, a podcast and live-audio application with approximately 900 daily active users across iOS and Android.

  • Developed a daily Sudoku feature in C#/.NET MAUI with XAML and MVVM, including notes, conflict detection, undo/redo, and completion validation.
  • Built an authenticated ASP.NET Core puzzle API, using Quartz.NET scheduling and randomized backtracking to generate daily boards.
  • Added 24-hour IMemoryCache caching and SemaphoreSlim synchronization to prevent duplicate loads. Cached puzzle retrieval fell from 34 ms to 0.2 ms in benchmarks.

03 / THE PERSON BEHIND THE CODE

Curious about
what’s underneath.

I like knowing what happens after you click the button. The API call, the database query, the background job that quietly keeps things moving — that’s usually where my curiosity takes me.

I’m a computer science student at Queens College, building across backend systems, full-stack applications, and data engineering. I enjoy getting a first version working, then figuring out what makes it dependable. Lately, that’s meant thinking more about data correctness and where applied AI can actually be useful.

A small detail outside the code

I’m also a One Piece fan. There’s always room for one more chapter.

The tools I work with

Chosen for the problem at hand.

Languages
Python · Java · SQL · JavaScript · TypeScript · C#
Backend
Spring Boot · Spring Security · ASP.NET Core · Node.js · Express
Frontend
React · Tailwind CSS · TypeScript
Databases & data
PostgreSQL · PostGIS · MySQL · Redis · SQLAlchemy
AI integration
Gemini API
Testing & tools
pytest · JUnit · Testcontainers · GitHub Actions · Docker · Git · Linux · OCI

04 / KEEP IN TOUCH

Have something
worth building?

I’d like to hear about it. For opportunities,
questions, or a conversation about software: