Open to ML & Data Roles

Md Shahedul
Islam Khan

$ ML Engineer · Data Scientist · AI Researcher

Building pipelines, models, and intelligent systems that work in the real world. Published researcher with industry project delivery and a passion for turning messy data into working systems.

View Projects Get in Touch ↓ Download CV Beyond Work →
Shahedul
600+
Students Taught
100+
Projects Supervised
2
Publications
5+
ML Systems Built
// about me

Research rigour.
Real-world delivery.

I'm an ML and data professional based in Newcastle, Australia, with end-to-end experience building pipelines, analytical systems, and predictive models. I hold an MSc in Cybersecurity from the University of Newcastle and a BEng in Computer Science from Northwestern Polytechnical University, China. Currently working as a Research Assistant on bias detection in healthcare AI, while teaching and mentoring in data-intensive computing courses.

// technical stack

The stack.

Languages
PythonSQLLaTeX
ML & Libraries
PyTorchScikit-learnHuggingFacePandasNumPy
Data & Analytics
Apache SparkMySQL ServerPower BI
Concepts
NLPComputer VisionLLMsPredictive ModelingETL Pipelines
Tools & Platforms
GitLinuxAWS
// experience

Where I've worked.

Research Assistant — NLP & Bias Detection
Current
University of Newcastle · Newcastle, NSW
  • Building BERT-family and LLM-based classifiers for multi-class bias detection and mitigation in healthcare text
  • Engineered scalable data pipelines with stratified sampling, proxy signal scoring, and statistical validation
  • Iterative model refinement using fairness metrics and explainability auditing
Associate Lecturer / Casual Academic
May 2024 – Present
University of Newcastle · Newcastle, NSW
  • Delivered lab instruction to 600+ students covering Big Data, database management, and data wrangling
  • Personally supervised 100+ capstone projects with technical mentorship and assessment
  • Marked and evaluated technical assignments delivering structured feedback
Undergraduate Research Assistant — ML & NLP
Oct 2021 – Jun 2022
Northwestern Polytechnical University · Xi'an, China
  • Designed an end-to-end ML pipeline for Chinese text classification
  • Research resulted in a peer-reviewed publication at ICIC 2022 (Springer)
// projects

Things I've built.

Industry-Affiliated · Tomago Aluminium
Phantom Signal — OSINT Social Engineering Simulation
End-to-end ML pipeline synthesizing realistic personas from OSINT data using transformer-based NLP. Demonstrated real-world organisational security vulnerabilities.
Computer Vision · YOLOv7
Automated Rock Detection — Industrial CV Pipeline
Production-ready object detection using YOLOv7 for rock identification in coal conveyor imagery. Handles variable lighting, occlusion, and high-speed conditions.
LLMs · Prompt Engineering · RAG
Intelligent Library Assistant
Domain-specific LLM chatbot with custom fine-tuning and prompt engineering. Context-aware retrieval and response generation for university library services.
Deep Learning · Agriculture
Multi-Modal Image Translation System
Deep learning model converting greyscale agricultural imagery to simultaneous infrared and RGB outputs for crop health monitoring.
Transfer Learning · NLP
Domain-Adaptive Sentiment Classifier
Fine-tuned BERT for multi-domain sentiment classification with targeted fine-tuning and layer freezing strategies.
// research

Publications.

arXiv 2024 · Cryptography & Security
A Data-Driven Predictive Analysis on Cyber Security Threats with Key Risk Factors
Fatama Tuz Johora, Md Shahedul Islam Khan, et al. · arXiv:2404.00068
ICIC 2022 · Springer · Pages 170–182
An Effective Chinese Text Classification Method with Contextualized Weak Supervision for Review Autograding
Yupei Zhang, Md Shahedul Islam Khan, et al.
// contact

Let's work together.

Open to ML, Data Science, and AI Research roles. Always keen to connect.

mdshahedulislam.khan@newcastle.edu.au