Syahmi Shamsul

Syahmi Shamsul

// Data Science & AI

Computer Science student at Taylor's University specialising in Data Science and AI. I build ML models, analytics dashboards, and turn raw data into actionable insights.

View Projects ↓ GitHub ↗ LinkedIn ↗
syahmi.py
1class SyahmiShamsul:
2  def __init__(self):
3    self.name = "Syahmi Shamsul"
4    self.role = "Data Science & AI"
5    self.university = "Taylor's University"
6    self.skills = ["Python", "ML", "DL"]
7    self.open_to_work = True
8 
9  def get_accuracy(self):
10    # best model score so far
11    return 93.58%
01 — About Me

Who I Am

7+
ML Projects
93%
Best Accuracy
2026
Graduating

I'm a final-year Computer Science student at Taylor's University, specialising in Data Science with an extension in Artificial Intelligence. I'm passionate about building models that solve real problems — from detecting bone fractures in X-rays to predicting loan defaults.

Beyond the models, I care about the full pipeline: clean data, meaningful insights, and dashboards that tell a story. My capstone project is MediVision — a full-stack Smart Patient Flow Platform for hospital emergency departments.

I'm passionate about turning complex data challenges into impactful solutions, and I'm always looking for new ways to push the boundaries of what's possible with data.

UniversityTaylor's University
Emailsabin.shamsul@gmail.com
GitHubgithub.com/sabinshamsul
LinkedInlinkedin.com/in/sabinshamsul
LanguagesEnglish (Advanced), Malay (Native)
02 — Skills

Tech Stack

Programming
Python R SQL Java pandas NumPy scikit-learn
Machine Learning
Supervised Learning Unsupervised Learning Classification Regression Clustering Random Forest kNN Decision Tree
Deep Learning
TensorFlow PyTorch LSTM CNN Neural Networks
Tools & Platforms
Jupyter Google Colab VS Code PyCharm Tableau GitHub Gradio Streamlit
03 — Projects

What I've Built

🏥
Final Year Project

MediVision — Smart Patient Flow Platform

MediVision is a full-stack hospital emergency department patient flow management system built to streamline hospital pathway, triage and clinical decision-making in real time built by a team of 5 members for our final year project (Capstone Project).

React Node.js Express.js MongoDB Tailwind CSS Vite
Live Demo
🧠
Deep Learning

Mental Health Sentiment Analysis

Built two text classification models on 53K+ mental health statements and compared Decision Tree vs Neural Network performance.

93.58% Neural Network · 90.78% Decision Tree
Python scikit-learn Gradio TF-IDF MLP Hugging Face
Live Demo
🎬
Deep Learning

Sentiment Analysis on IMDB Reviews

Performed sentiment analysis using a Bidirectional LSTM model on IMDB movie reviews in Google Colab. Binary classification of positive vs negative reviews.

86.8% Accuracy — BiLSTM
Python TensorFlow LSTM
Live Demo
💊
Machine Learning

Analyse Lethal Drug Combinations

Built data mining models on drug-death records. Applied Random Forest for classification, K-Means for seasonal clustering, and Isolation Forest to detect geographic outliers.

81% Accuracy — Random Forest
Python scikit-learn K-Means
Live Demo
🏦
Machine Learning

Predicting Loan Defaults

Developed ML models to analyse loan default risk on a financial dataset. Compared k-Nearest Neighbours against Decision Tree classifiers on key financial features.

88.54% kNN · 88.07% Decision Tree
Python kNN Decision Tree
Live Demo
🦴
Computer Vision

Bone Fracture Detection

Developed a Convolutional Neural Network (CNN) to automatically detect bone fractures in X-ray images. Trained on labelled radiography data with data augmentation techniques.

73.84% Accuracy — CNN
Python CNN TensorFlow
Live Demo
🌿
IoT

Smart Monitoring for Houseplants

Built an IoT prototype using ESP32 and Raspberry Pi with sensors to automate plant watering and monitor moisture, temperature, humidity, and soil pH in real time.

ESP32 Raspberry Pi Python
GitHub ↗
❤️
Machine Learning

Heart Disease Prediction

Conducted EDA and preprocessing on the UCI Heart Disease dataset using Python and Scikit-learn. Developed a Logistic Regression model to predict heart disease risk.

66.67% Accuracy — Logistic Regression
Python scikit-learn EDA
Live Demo
04 — Experience

Work & Activities

Student Assistant — FitFuel Station Research Project
Jan 2026 – Apr 2026
Taylor's University, Malaysia · INNOVATE Grant
  • Coordinating the 'FitFuel Station — An AI-Driven Vending Device for Foodstuff Customisation' research project
  • Assisting in software logic translation, data handling, and validating nutrition algorithms
  • Ensuring accurate integration between body composition inputs, decision engines, and AI-controlled dispensing output
UG Research Idea Pitch Competition Participant
Dec 2025
Research & Innovation Festival (RNIF 2025) · Taylor's University, Malaysia
  • Participated in the undergraduate research idea pitch competition at the university-wide innovation festival
05 — Education

Academic Background

Bachelor of Computer Science (Honours)
Taylor's University, Malaysia · Feb 2024 – Aug 2026 (Expected)
Specialisation: Data Science Extension: Artificial Intelligence
Capstone Project: MediVision — Smart Patient Flow Platform for Hospital Emergency Departments (Ongoing)
06 — Contact

Let's Connect

If you're building something interesting with data or AI, I'd love to hear from you. Let's connect.

Email
sabin.shamsul@gmail.com
→
LinkedIn
linkedin.com/in/sabinshamsul
→
GitHub
github.com/sabinshamsul
→