GitHub portfolio

HamzaaAkmal

AI Engineer & Full-Stack Developer building open-source tools with Python, TypeScript, and ML
Pakistan · Downlabs
Hamza Akmal
01 / About

Building with intent.

I'm Hamza Akmal, an AI Engineer and Full-Stack Developer based in Pakistan. I'm currently a DS student at Superior University and work at Downlabs, shipping 59 public repositories that span browser extensions, ML utilities, computer vision prototypes, and web applications. Recent open-source work includes an AI-powered Chrome extension for Islamic content filtering, a Python proxy that routes Claude Code requests across free and local LLM providers, and a contactless scrolling prototype using OpenCV and MediaPipe. I build with Python, TypeScript, JavaScript, Java, Kotlin, Swift, and PHP, and I’m focused on practical AI interfaces and accessible digital tools.
02 / Selected work

Projects

A selection of public work, shaped from live GitHub data.
TypeScript
★ 1
Aladdin-Video-Editor

Browser-based video editor as an open-source alternative to Capcut and InShot.

Built with TypeScript, Python, PL/pgSQL, and Docker to power in-browser video editing tools without requiring desktop installation.

browser-basedmediaopen-sourcetypescriptvideo-editor
Blade
★ 1
SEO-Tools-Script

SEO toolkit for website optimization and rank tracking.

Uses PHP, Blade templating, JavaScript, and CSS within the Laravel ecosystem to deliver web-optimization and analysis tools.

laravelphpseotoolsweb-optimization
HTML
★ 1
Crypto-Profit-Calculator

Real-time cryptocurrency profit calculator.

An HTML and JavaScript web tool that calculates investment profits with live price updates.

calculatorcryptocurrencyfinancehtmljavascript
Jupyter Notebook
★ 1
Fraud-Job-Detection-Model

ML system for identifying fraudulent job postings.

Applies NLP with TF-IDF vectorization and a LinearSVC classifier using Jupyter Notebook and Python, deployed on Streamlit for interactive fraud detection.

data-preprocessingdata-visualizationfeature-engineeringfraudlinearsvc
Swift
★ 0
CrisisX-Hackathon

Rapid-response hackathon build using Swift and TypeScript.

Combines Swift, TypeScript, and PL/pgSQL into a crisis-focused hackathon application.

Java
★ 0
SnapCARE

Multi-language mobile and backend codebase.

Developed using Java, Kotlin, Python, and PL/pgSQL across a shared repository.

03 / Capabilities

Tools & languages

Repository language usage and evidence-backed skills.
Language graph
JavaScript43%
Blade14%
PHP13%
TypeScript12%
Python7%
Jupyter Notebook4%
Java4%
Swift3%
Skill map
Languages & Markup
PythonTypeScriptJavaScriptJavaKotlinSwiftPHPHTMLCSSSQL
AI / ML & Data
Machine LearningNLPfeature engineeringscikit-learnTF-IDFLinearSVCOpenCVMediaPipeNumPyJupyter NotebookStreamlit
Systems & Tooling
Chrome Extension APIsDockerLaraveluvRuffPytestLoguruPyAutoGUI
Integration & Architecture
API proxy designreal-time web filteringheuristic tool parsingrate-limiting strategylocal LLM inferenceDiscord bot integrationTelegram bot integration
04 / Resume

Experience

The complete uploaded resume, available to read or download.