About

I am a computer scientist from Bangladesh, now a Ph.D. candidate in Computer Science and Engineering at the University of North Texas, advised by Dr. Renee Bryce.

I spent about four and a half years as a software engineer in Dhaka, leading Android teams, before I came to graduate school. That work left me curious about how software teams actually get things fixed, and it is still the thread that runs through what I do now.

  • Research

    I apply machine learning and natural language processing to software engineering problems, mostly automated bug triage. My work is published at SEKE, ITNG and IEEE CCWC, where a paper I co-authored won a Best Paper Award.

  • Engineering

    I have led mobile teams at two companies and shipped apps used in field programs across Africa and Asia, plus a video consultation platform.

  • Teaching

    Since 2022 I have worked alongside faculty in eight undergraduate and graduate courses at UNT, and I mentor undergraduate researchers.

Path

Study
B.Sc., AIUB Graduate study, UNTM.S. 2025, Ph.D. expected 2027
Work
Field BuzzAndroid team lead HelloDoc24Android lead University of North TexasTeaching and research
Milestones
Smart City Hackathon, top 5 Google Associate Android Developer First paper, ITNG Best Paper, IEEE CCWC

Bars show roles and study over time, fading out where they continue. Dots mark single events.

Research

Large software projects receive thousands of bug reports, and someone has to decide who should fix each one. When that call is wrong, the report gets passed from developer to developer and the fix waits. My dissertation builds models that make this routing decision from the report text, the project's records about components and teams, and its commit history. I test them the way they would be used in practice: trained on the past and evaluated on reports that come later.

  1. 2025

    Commit messages as a signal

    Adding developers' commit messages to an XLNet model raised Top-1 accuracy on Firefox from 23.28% to 69.45%, and on Eclipse from 47.89% to 84.88%.

  2. 2026

    A model a project could deploy

    A hybrid transformer combining report text with project metadata reached 69.02% Top-1 and 90.69% Top-10 accuracy on 25,220 Firefox reports. Text and metadata alone reach 99.6% of the full model, so it still works when no commit history exists yet.

  3. Next

    Routing that keeps up with real teams

    Ranking models that learn how reports actually move between developers, and that handle newcomers, changing teams, workload and signals from the code itself.

Other interests: information retrieval, graph neural networks, learning to rank, mobile and context-aware systems, and software testing. Experiments run on the Lonestar6 supercomputer at the Texas Advanced Computing Center through an NSF ACCESS allocation I wrote with Dr. Bryce.

Publications

  1. Enhancing Bug Triaging with Organizational Metadata and Commit-Aware Transformers F. R. Arnob, P. Goyal, M. K. Khan, R. Bryce 38th International Conference on Software Engineering and Knowledge Engineering (SEKE 2026). Regular paper, 41% acceptance rate. doi:10.18293/SEKE2026-010
  2. Hybrid Sequential Pattern Mining and Machine Learning for Context-Aware Event Prediction in IoT P. Goyal, B. Helmholz, T. Tran, F. R. Arnob, R. Bryce 2026 IEEE 16th Annual Computing and Communication Workshop and Conference (CCWC), Las Vegas, pp. 917–923. Best Paper Award (Reviewers' Choice) doi:10.1109/CCWC67433.2026.11393795
  3. Bug Triaging Based on Transformer Models Utilizing Commit Messages F. R. Arnob, R. H. Mollik, P. Goyal, R. Bryce 22nd International Conference on Information Technology: New Generations (ITNG 2025), Springer, pp. 354–366.
  4. Prioritization Techniques for Android Test Suites Generated by a Reinforcement Learning Algorithm M. K. Khan, R. Michaels, F. R. Arnob, R. Bryce Preprint, SSRN. doi:10.2139/ssrn.4450321

Citations and updates on Google Scholar.

Experience

  • 2022–present
    Graduate Teaching Assistant and Researcher, University of North Texas

    Research in Dr. Renee Bryce's software engineering lab; teaching support across undergraduate and graduate courses in Computer Science and Engineering.

  • 2021–2022
    Senior Software Engineer and Android Lead, HelloDoc24, Dhaka

    Led the Android team for a video consultation app with role-based access, built in Kotlin with Jetpack, coroutines and Azure Communication Services.

  • 2018–2021
    Software Engineer and Android Team Lead, Field Information Solutions (Field Buzz), Dhaka

    Became team lead on three projects in my first year. Built field data apps for organizations including BRAC Microfinance Rwanda and UNDP, and for projects in Mali such as Projets Eau and Pistes pour la Paix. Wrote shared camera, login and utility modules reused across products.

Education

  • 2027 (expected)
    Ph.D., Computer Science and Engineering

    University of North Texas. Advisor: Dr. Renee Bryce.

  • 2025
    M.S., Computer Science

    University of North Texas

  • 2018
    B.Sc., Computer Science and Engineering

    American International University–Bangladesh

Teaching and mentoring

Most of my teaching has been in IT Project Management, a course of 50 to 100 students who go straight on to their senior capstone. I help build the syllabus, assignments and quizzes, run the Canvas course, and work with student teams on planning and leadership, drawing on my own years as a team lead.

  • CSCE 3055 IT Project Management, 11 offerings
  • CSCE 3220 Human Computer Interfaces
  • CSCE 3530 Introduction to Computer Networks
  • CSCE 4901 / 4905 Capstone sections
  • CSCE 5210 Fundamentals of Artificial Intelligence
  • CSCE 5270 Computer-Human Interfaces
  • CSCE 5900 Special Problems
  • CSCE 5934 Directed Study

Mentoring

I mentor undergraduates through the NSF-funded TaMaLe REU site at UNT; one of those projects became the CCWC 2026 Best Paper. At Field Buzz I ran training sessions for my team on design patterns, clean code, reactive programming and Material UI.

Skills

Languages
Python, Java, Kotlin
Machine learning
Transformer models (RoBERTa, CodeBERT, DeBERTa-v3, XLNet, T5), LightGBM LambdaMART, TensorFlow, scikit-learn, NLTK, Pandas, LLMs
Mobile
Android (Jetpack, MVVM, coroutines and Flow, Retrofit, WorkManager), Espresso and unit testing
Platforms
TACC Lonestar6 (HPC), Azure, AWS SageMaker, WebRTC, Android Studio

Honors and service

  • 2026
    Best Paper Award (Reviewers' Choice), IEEE CCWC
  • 2025
    Excellent Program Committee Member, IEEE ICMLA
  • 2025
    HPC allocation, TACC and NSF ACCESS
  • 2025
    Finalist, Innovate UNT Pitch Competition

    Technical lead for team VeriWell.

  • 2024
    International Education Scholarship and CEO Prep Award, UNT
  • 2017
    Google Associate Android Developer
  • 2016
    Top 5, Smart City Hackathon Bangladesh

    Team lead.

Service and community

  • 2025–2026
    Program committee member

    IEEE ICTAI 2026, IEEE ICMLA 2026 and IEEE ICMLA 2025.

  • 2024–2025
    Judge

    Eagle Awards at UNT (2025) and the TAMS Science Fair (2024).

  • 2024–2025
    General Secretary, Bangladesh Student Association, UNT