Software Engineer | MS Computer Science, Colorado State University

Masfiq Reza

Builds tested backend and data software in Python and C#: REST APIs, concurrency safe transactions, and ETL pipelines with automated validation.

View Projects Download Resume GitHub

Pharmacy Order Fulfillment ServiceDetails

ReceivePrescription orders arriveEmergencyRoutine
ScheduleHeap based priority scheduler
FillAtomic conditional stock updates prevent overselling
AuditEvery state transition logged in the same transaction

~330orders per second

<108 msp95 latency

98%coverage, 102 tests

Drug Lookup and Inventory APIDetails

LoadIdempotent ETL of FDA drug records into PostgreSQL
Validate5,366 data quality issues logged with reasons, none silently dropped
SearchTrigram indexes for typo tolerant name search488.8 ms4.7 ms
DispenseFirst expiring, first out with row locks in transactions

138,427FDA drug records

104×faster fuzzy search

97%coverage, 87 tests

Pharmacy Inventory API in C#Details

ApiREST endpoints with ProblemDetails errorsASP.NET Core
CoreLot dispensing as a pure function, free of database code
DataRepository interfaces over EF Core and SQLite
TestxUnit unit and integration tests run in GitHub Actions

59xUnit tests

95.8%line coverage

3layers: Api, Core, Data

Colorado Inequality Data HarmonizerDetails

IngestFive federal sourcesACSHMDAUSDACDC PLACESGazetteer
TransformETL in Python and Pandas, 2010 and 2020 tracts reconciled
Validate23 automated validation checks
ServeFastAPI backend to a Streamlit dashboard map

1,447census tracts

130columns

27%redrawn tracts kept

Projects

Software I have built

Data pipelines, APIs, web applications, and deep learning systems, written mostly in Python.

Backend

Pharmacy Order Fulfillment Service

A central fill pharmacy backend with priority scheduling, safe concurrent stock updates, and a full audit trail.

Python
FastAPI
PostgreSQL
Docker

Built a central fill pharmacy backend with a heap based priority scheduler and an order state machine whose every transition is written to an audit log in the same database transaction.
Prevented overselling with atomic conditional stock updates and database constraints; a test firing 20 simultaneous requests at stock for 7 orders yields exactly 7 fills and zero negative stock.
Load tested at about 330 orders per second (p95 under 108 ms); in a simulated 8 hour day, emergency orders waited about 18 times less than routine ones; 102 tests at 98% coverage run in GitHub Actions.

Backend and data

Drug Lookup and Inventory API

A FastAPI service over 138,427 FDA drug records with fast, typo tolerant search.

Python
FastAPI
PostgreSQL
Pandas

Built an idempotent ETL pipeline loading 138,427 FDA drug records into a normalized PostgreSQL schema, logging all 5,366 data quality issues with reasons instead of silently dropping rows.
Added typo tolerant search with trigram indexes, cutting fuzzy name search from 488.8 ms to 4.7 ms (104 times faster); other indexes sped up ingredient and expiry queries 33 and 82 times.
Implemented first expiring, first out dispensing with row locks inside transactions; 87 tests at 97% coverage.

C# and .NET

Pharmacy Inventory API in C#

A layered ASP.NET Core REST API for pharmacy inventory with lot based dispensing.

C#
ASP.NET Core
EF Core
SQLite
xUnit

Built a layered REST API (Api, Core, Data) using dependency injection and repository interfaces, keeping business rules free of database code.
Wrote lot dispensing as a pure function inside transactions, with ProblemDetails errors and a daily expiry alert.
59 xUnit unit and integration tests (WebApplicationFactory) at 95.8% line coverage, run in GitHub Actions.

Data engineering

Colorado Inequality Data Harmonizer

One validated, analysis ready table of Colorado census tract data built from five federal sources.

Python
FastAPI
Streamlit
Pandas
GeoPandas

Built an ETL pipeline merging five federal sources (ACS, HMDA, USDA Food Access Atlas, CDC PLACES, Census Gazetteer) into one 130 column table covering all 1,447 Colorado census tracts.
Wrote 23 automated validation checks and a generated data dictionary; reconciled 2010 and 2020 tract boundaries with documented fallbacks instead of dropping the 27% of redrawn tracts.
Served the versioned feature set through a FastAPI backend to a Streamlit dashboard with a live map.

MS Thesis

Aboveground Biomass Estimation

A deep learning regression pipeline that estimates forest biomass from sparse LiDAR labels and satellite data.

Python
PyTorch
SLURM
GeoPandas

Building a deep learning regression pipeline fusing sparse NASA GEDI LiDAR with Harmonized Landsat Sentinel imagery, Sentinel 1 radar, and gridMET climate data.
Running data processing and model training jobs on CSU's Falcon HPC cluster with SLURM and Bash.

Deep learning

Alzheimer's Disease Classification from MRI

A custom 3D CNN that classifies Alzheimer's disease from MRI volumes.

Python
PyTorch

Designed a custom 3D CNN to classify Alzheimer's disease from MRI volumes in the ADNI dataset.
Published at IEEE TENCON 2021.

More projects

Shop Management Web Application

Inventory and finance management for shops with multiple vendors, built with Django.

Python
Django

Wildfire Risk Prediction

Feature engineering and model comparison under class imbalance.

Python
scikit-learn
Pandas

Experience

Where I have worked

Aug 2026 to Present
Graduate Research Assistant, AI Ready Inequality Data Hub
Colorado State UniversityFort Collins, CO
- Building Python ETL pipelines that clean, join, and validate federal socioeconomic data for Colorado census tracts.
Sep 2026 to Present
Research Computing Support Technician
Colorado State University, Office of the Vice President for ResearchFort Collins, CO
- Helping researchers get started on CSU's HPC systems and improve their Linux, SLURM, Python, and Bash workflows.
- Supporting test driven development of HPC tools and sponsored research projects that need computing and storage services.
Aug 2024 to May 2026
Graduate Teaching Assistant
Colorado State UniversityFort Collins, CO
- Supported labs and recitations for Operating Systems, Modern Cybersecurity, and Modern Web Development.
- Wrote autograder scripts and gave detailed, rubric based feedback on assignments and exams.
- Held office hours debugging student programs in C, C++, and Python on Linux, and taught Git workflows.
Feb 2023 to July 2024
Lecturer, Computer Science
Uttara UniversityDhaka, Bangladesh
- Taught undergraduate CS courses, built course materials, and graded assignments and exams.
Nov 2022 to Feb 2023
Trainee Software Engineer (SQA Automation)
BJIT LimitedDhaka, Bangladesh
- Built automated UI test suites in Java with Selenium and TestNG for web applications.
- Designed and executed test scenarios for API testing (Postman) and load testing (Apache JMeter).
Jun 2021 to Apr 2022
Student Tutor
BRAC UniversityDhaka, Bangladesh
- Taught Data Communications and introductory programming; graded and supported lab sections.

Skills

Tools I work with

Tip: grab a chip and give it a fling.

Languages

Python
C#
Java
C++
C
JavaScript
SQL
Bash
HTML
CSS

Backend and web

FastAPI
ASP.NET Core
Entity Framework Core
SQLAlchemy
Alembic
Django

Databases

PostgreSQL
MySQL
SQLite

Testing and DevOps

pytest
xUnit
Selenium
TestNG
Postman
JMeter
Git
GitHub Actions
Docker

Data and ML

Pandas
NumPy
scikit-learn
PyTorch
Parquet
ETL pipelines
Data validation
Linux

Geospatial and HPC

GeoPandas
rasterio
NASA GEDI LiDAR
HLS imagery
Sentinel 1 SAR
SLURM

Education

Degrees

GPA 3.92/4.00

Master of Science in Computer Science

Colorado State University

Fort Collins, COExpected May 2027

Coursework: Natural Language Processing, Artificial Intelligence

CGPA 3.80/4.00

Bachelor of Science in Computer Science and Engineering

BRAC University

Dhaka, Bangladesh2018 to 2022

Coursework: Data Structures, Algorithms, Object Oriented Programming, Database Systems

VC's List (3 times)
Dean's List (5 times)
University Merit Scholarship and Stipend

Writing and research

Five papers, including WACV 2027 (accepted), ACM/IEEE ESEM 2025 (LLM based task recommendation for open source software), Data in Brief 2024, and IEEE TENCON 2021.

Medium tutorials on NASA GEDI LiDAR, Harmonized Landsat Sentinel imagery, and the ADNI dataset.

Google Scholar Medium

Get in touch

Looking for software engineering roles. Email is the fastest way to reach me.

Based in Fort Collins, Colorado