2024 in Review: Graduation, First Publications and a Classroom

2024 was the year my work moved from coursework to publication, and from studying computer science to teaching it. Often in the same week.

Writing a year in review is partly for other people and mostly for me. It forces an honest accounting of what actually got finished rather than what felt busy. Here is what 2024 amounted to.

Teaching

I spent the year as a Lecturer at Sunshine Grammar School and College in Chattogram, teaching Computer Science and Applied Information Technology across both the Western Australian and British curricula.

Teaching two curricula in parallel turned out to be the most useful thing that happened to me professionally. The same topic is framed differently in each, assessed differently, and weighted differently, so preparing a lesson means understanding the material twice over rather than reciting it once. Several concepts I thought I had learned properly during my degree did not survive contact with a class of students asking why.

The part I did not anticipate was how much of the job is mentoring rather than instruction. A capable student who is convinced they are not capable needs something quite different from a better lesson plan. I got noticeably better at that over the year, and it is the skill I most want to keep developing.

Graduation and research

I completed my undergraduate degree in Computer Science in 2024. My final-year project was a credit card fraud detection system built with machine learning, which means training a model to flag the small number of fraudulent transactions hidden inside millions of legitimate ones.

The interesting part of that problem is the imbalance. Fraud is rare, so a model that simply predicts “not fraud” every single time scores over 99 percent accuracy and is completely useless. Getting anything meaningful out of it meant learning to care about the right measures rather than the flattering one, and to weight the cost of a missed fraud against the cost of a false alarm.

That project became my first research paper, CLocMLGuard: A Novel Risk Scoring Framework for Transaction Card Fraud Mitigation and Detection with Machine Learning, accepted at the 16th IEEE International Conference on Computing Communication and Networking Technologies.

Alongside it, work I contributed to during the year has since appeared at IEEE and Springer venues:

• Hybrid Machine Learning Model for Detecting Bangla Smishing Text Using BERT and Character-Level CNN, International Conference on Electrical and Computer Engineering (ICECE), IEEE, 2024.
• PhishVQC: Optimizing Phishing URL Detection with Correlation Based Feature Selection and Variational Quantum Classifier, International Conference on Intelligent Systems, Advanced Computing and Communication (ISACC), IEEE, 2025.
• Quanvolutional Neural Networks for Pneumonia Detection: An Efficient Quantum-Assisted Feature Extraction Paradigm, 2nd International Conference on Next-Generation Computing, IoT and Machine Learning (NCIM), IEEE, 2025.
• Quantum Machine Learning for Image Classification: A Hybrid Model of Residual Network with Quantum Support Vector Machine, NCIM, IEEE, 2025.

The Bangla smishing paper is the one I am most attached to, because smishing, which is phishing delivered by SMS, is a problem that looks solved in English and is barely addressed in Bangla. Building for a language that most models were never trained on is a different kind of research problem, and one that matters directly to people here.

Behind all of it was the unglamorous work of getting fluent with NumPy and Pandas, the Python libraries that handle numerical arrays and tabular data and that almost every research pipeline is built on. Certificates prove nothing on their own. Being able to write the code without looking things up every ten minutes is what made the papers possible.

Building a research presence

I built this website in 2024, for a straightforward reason. Research nobody can find does very little, and a scattered trail across LinkedIn, ORCID, Google Scholar and ResearchGate is not a substitute for one page that holds the publications, the teaching record and the ongoing projects together.

It has been more useful than I expected, mostly through conversations that started because somebody found a paper here first.

Outside the work

Two things from the year stand out.

In January my family finished building our house in the village. That project ran for years before it was done, and watching it complete was the kind of milestone that does not fit anywhere on a CV and matters more than most of the things that do.

In April we travelled through India together, covering Kolkata, Delhi, Hyderabad, Rajasthan and Mumbai. Five cities in one trip is too many to see any of them properly, which I would argue is the point. What stayed with me was the range rather than the monuments. Hyderabad and Rajasthan felt like different countries, and the ordinary kindness we met from strangers was the one thing consistent across all five.

My English improved substantially over the year, mostly through teaching. Explaining a technical idea to a room of students who will interrupt the moment you are unclear is the most efficient language practice I have found.

What I am working towards

The direction from here is a funded PhD position abroad, researching machine learning and its applications to security.

That is a target rather than an aspiration, which means it comes with specific work attached. More papers, and stronger ones, in better venues. Real depth in the areas I want to research instead of broad familiarity across many. A teaching record that holds up when somebody actually reads it rather than skims it.

2024 was the year the foundation went in. The building comes next.

---

My full publication list is on the research page, on Google Scholar and on ORCID. If you are working on something related, the contact page is the fastest way to reach me.