ML Engineer & Researcher

M Muzammil
Irshad

Building Intelligent Systems · Advancing AI Research

A driven Computer Science student at University of Education, Lahore — engineering machine learning systems at the intersection of deep learning, computer vision, and cybersecurity research.

MI
3.66
CGPA
5+
Projects
6th
Semester
About

Crafting Intelligence
From Data

I am a 6th semester BS Computer Science student at the University of Education, Lahore — Vehari Campus, with a strong academic record and a clear, ambitious trajectory toward becoming a Machine Learning Engineer.

My journey spans classical ML, deep learning, and computer vision — with hands-on projects already built and more in active development. But my vision does not stop there. I am steadily advancing toward the frontier of Agentic AI — building systems that don't just predict, but reason, plan, and act autonomously.

Beyond coursework, I am actively conducting research in cybersecurity — applying adaptive machine learning to detect insider behavioral threats under real-world conditions of concept drift. I believe intelligence, when engineered responsibly, makes both systems and societies safer.

Institution
University of Education, Lahore
Campus
Vehari Campus
Degree
BS Computer Science
Semester
6th Semester
CGPA
3.66
Academic Excellence
Projects

Systems Built
With Purpose

002
Computer Vision
Face Mask Detection

Real-time face mask detection using deep learning, identifying masked and unmasked individuals from live video streams and static images.

Deep LearningOpenCV
003
Cybersecurity · ML
Network Intrusion Detection

ML-powered model for detecting anomalous network traffic and classifying intrusion types — supporting proactive cybersecurity threat response.

In Progress
Scikit-LearnSecurity
004
NLP · Classification
Email Classifier

Intelligent email classification using NLP to categorize emails with high accuracy, filtering spam and organizing communication pipelines.

NLPTF-IDFPython
005
NLP · Media
News Classifier

Automated news article classification system categorizing content into relevant topics using machine learning and text feature engineering.

TF-IDFScikit-Learn
Skills

Technical Arsenal

Core Skills
Python100%
Machine Learning90%
Deep Learning50%
NLP30%
Languages
PythonSQLBash
Data Analysis
NumPyPandasMatplotlibSeaborn
Machine Learning
Scikit-LearnXGBoostFeature EngineeringModel Evaluation
Deep Learning
TensorFlowKerasTransfer LearningCNN
NLP
NLTKTF-IDFText Classification
Computer Vision
OpenCVImage ProcessingObject Detection
Tools & Platforms
Google ColabJupyterGit & GitHubKaggleVS Code
Research

Active Research
Pursuits

Domain — Cybersecurity × Machine Learning
Adaptive and Incremental Approaches to Detect Behavioral Insider Threats Under Concept Drift

This research investigates how machine learning models can be made adaptive and incrementally updatable to detect insider behavioral threats in dynamic environments — particularly under concept drift, where the statistical properties of data shift over time. Traditional static models fail in such settings. The work explores drift detection algorithms, online learning, and incremental model updates to ensure sustained detection accuracy against evolving insider threat patterns — contributing to behavioral analytics, adaptive ML, and enterprise cybersecurity.

Status
Active — Ongoing
Domain
Cybersecurity
Focus
Insider Threat Detection
Methods
Adaptive ML · Concept Drift

Let's Connect

Open to collaborations, research discussions, and opportunities in ML Engineering. Let's build something meaningful.