Skip to content

Lindiwe-22/Portfolio-Website

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

88 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Lindiwe Songelwa — Data Science & DevOps Portfolio

A fully responsive, single-page portfolio website built with pure HTML, CSS, and JavaScript — showcasing 12 projects across two disciplines: Data Science and DevOps Engineering (AZ-400).

🔗 Live Site: lindiwe-22.github.io/Portfolio-Website


📁 Project Structure

portfolio-website/
│
├── index.html          # Main HTML structure — all sections and project cards
├── styles.css          # All CSS styles, animations, and responsive layout
├── script.js           # JavaScript for interactivity and scroll behaviour
└── README.md           # Project documentation

🗂️ Featured Projects

🔬 Data Science

Project Description Stack
SA Crime Intelligence Report Rigorous interrogation of Q3 2025/26 SAPS crime statistics cross-referenced with SAMRC femicide data. Applies OLS regression, ARIMA forecasting, correlation analysis and anomaly detection. Python, Statsmodels, Scikit-learn, ARIMA
Mining Quality Intelligence Full-stack ML system predicting silica concentrate failures in an iron ore flotation plant, trained on 737,453 sensor readings with a live GREEN/AMBER/RED Streamlit dashboard. XGBoost, TensorFlow, SMOTE, Streamlit, Plotly
Tyla Grammy Sentiment Analysis NLP study revealing 82% positive sentiment across 214 fan reactions in 20+ countries following Tyla's historic Grammy win. Dual-model validation with VADER and TextBlob. Python, VADER, TextBlob, YouTube API, WordCloud
Diamonds Are Forever Full-stack ML investigation into the diamond industry — quantifying the 73.8% lab-grown price collapse, K-Means buyer archetypes, and price forecasting at R²=0.9997. Random Forest, XGBoost, K-Means, Streamlit, Plotly
SA Bank Trust Score Consumer intelligence dashboard scoring South Africa's six major banks across complaint resolution, regulatory sanctions, consumer favour rate and public sentiment. Python, Streamlit, Pandas, Matplotlib, OBS/SARB Data
Who Really Gets the Money? Public interest investigation into R69.5 billion of SA development finance — combining Gini/Lorenz inequality analysis, ML cost-per-job prediction, and Isolation Forest anomaly detection. Scikit-learn, XGBoost, Isolation Forest, Streamlit, Plotly

⚙️ DevOps (AZ-400 Portfolio)

Project Description Stack
SA Bank Trust Score — DevOps Pipeline Production-grade CI/CD pipeline with four GitHub Actions workflows: app health checks, CSV data quality validation, scheduled data freshness monitor with auto Issue creation, and Docker build verification. GitHub Actions, Docker, CI/CD, YAML
Tyla Sentiment Tracker Automated fortnightly sentiment tracking pipeline — YouTube comments analysed on a scheduled GitHub Actions workflow with no manual intervention. GitHub Actions, VADER, TextBlob, Scheduled CI
Prometheus — Grafana Monitoring Production observability stack for the Iris Model API — metrics collection, dashboard visualisation and alerting in a fully containerised environment. Prometheus, Grafana, Docker
K8s Iris Deployment Production-grade Kubernetes deployment demonstrating container orchestration, autoscaling, and rolling updates using Kubernetes manifests and kubectl. Kubernetes, Docker, kubectl
Terraform — Docker IaC Infrastructure as Code project provisioning a full Docker stack using Terraform — multi-environment deployment, resource management and infrastructure versioning. Terraform, Docker, HCL
ML CI/CD Pipeline End-to-end ML model deployment pipeline built on GitHub Actions — automated training, testing, and deployment stages for a machine learning model. GitHub Actions, Python, CI/CD

🛠️ Technologies Used

Frontend

  • HTML5 — Semantic markup and accessible structure
  • CSS3 — Custom properties, CSS Grid, Flexbox, keyframe animations, glassmorphism
  • JavaScript (ES6+) — Intersection Observer API, scroll behaviour, dynamic panel toggling

Data Science Stack

  • Python, Pandas, NumPy, Scikit-learn, XGBoost, TensorFlow
  • Statsmodels (OLS, ARIMA), VADER, TextBlob
  • Streamlit, Plotly, Matplotlib, Seaborn

DevOps Stack

  • GitHub Actions (CI/CD workflows, scheduled jobs, artifact management)
  • Docker (containerisation, Docker Compose, multi-stage builds)
  • Kubernetes (manifests, kubectl, autoscaling, rolling updates)
  • Terraform (IaC, HCL, multi-environment provisioning)
  • Prometheus & Grafana (metrics, observability, alerting)
  • YAML (workflow and manifest authoring)

🎨 CSS Techniques

  • Keyframe animations — floating hero circles, skill bar shimmer, morphing profile border
  • CSS custom properties — centralised design tokens for colour, shadow, and gradient
  • Glassmorphismbackdrop-filter: blur() on the fixed navigation bar
  • Intersection Observer API — scroll-triggered animations on section entry
  • SVG illustrations — inline, self-hosted project card visuals (no external image dependencies)
  • Mobile-first responsive design — breakpoints at 768px and 480px

🚀 Deployment

The portfolio is deployed via GitHub Pages and updates automatically on every push to main.

Live URL: https://lindiwe-22.github.io/Portfolio-Website/

To run locally:

git clone https://github.com/Lindiwe-22/Portfolio-Website.git
cd Portfolio-Website
# Open index.html directly in any modern browser
start index.html       # Windows
open index.html        # Mac
xdg-open index.html    # Linux

📱 Responsive Breakpoints

Breakpoint Target
1200px+ Desktop
768px – 1199px Tablet
320px – 767px Mobile

📧 Contact

Lindiwe Songelwa


Built with HTML, CSS, JavaScript — and a lot of intention.

About

CSS website created specifically for my portfolio

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors