Beyond the Code
I am a Full Stack Developer based in Dubai, UAE with 4+ years of experience building mobile applications and backend systems for enterprise use.
I work across the full stack — from native mobile UI in Android, Flutter, and iOS, to backend APIs with Python (FastAPI · Flask), cloud deployments on AWS, and AI/LLM integrations. My journey started at COMSATS University (BS CS, 2019–2023) where I researched deep learning for medical imaging.
Mobile Expert
Android · Flutter · iOS across fintech & field service
Backend & Cloud
FastAPI · Flask · Docker · AWS · PostgreSQL
COMSATS Islamabad
Technical Arsenal
A full-stack toolkit spanning native mobile, Python backends, cloud infrastructure, and on-device AI.
Mobile Development
Backend & Cloud
AI & LLM
DevOps & Security
Featured Deployments
From native mobile modules to scalable D2C platforms.



Fiorra Bedding
A high-performance D2C e-commerce platform engineered for my own bedding brand.
Full-Stack Architecture
Built with Next.js App Router and Server Actions for SEO dominance. Custom backend handles order orchestration, inventory locking, and automated email logistics.
Business Logic & CMS
Developed a bespoke Admin Panel for real-time sales analytics and product management, eliminating monthly Shopify fees and allowing complete data ownership.
UX & Performance
Optimized core web vitals with image lazy-loading and edge caching.
Source code is proprietary. Live architecture walkthrough available upon request.
Contactless Biometrics
Enterprise banking module enabling secure finger verification via smartphone cameras. Achieved 99% acceptance accuracy and sub-200ms latency using custom TFLite quantization.
SpeechDriven Banking SDK
"Siri-like" conversational interface for fintech. Implemented full-duplex WebSockets with Mu-law compression (50% bandwidth reduction) and on-device VAD.
Offline-First FSM App
Utility app for 2000+ field workers. Features on-device OCR to digitize meter readings, reducing human error by 90%. Built with WorkManager for 100% data integrity.
LucaSeg: Cancer Detection
Academic research project using 3D CNNs for Lung Cancer segmentation on CT-scans. Features a real-time visualization dashboard for clinicians to classify nodules.
LLM RAG Pipeline
Personal project: a Retrieval-Augmented Generation pipeline integrating Ollama with the Llama model for document-based question answering using local LLM inference — no cloud API costs.
Professional Timeline
Full Stack Developer
Python · Mobile · Cloud
wAI Industries — Remote (Dubai Based)
- Developed and deployed RESTful APIs using FastAPI and Flask, containerized with Docker and hosted on AWS EC2, with S3 for media and document storage.
- Managed PostgreSQL database schema design and integration with backend API services.
- Designed and developed an Android application (Kotlin, Jetpack Compose, MVVM, Hilt) for field service operations, supporting 2,000+ field users with real-time job tracking via Google Maps SDK.
- Built an offline-first sync system using Room and WorkManager, ensuring data availability under poor or no connectivity.
- Integrated TensorFlow Lite OCR to automate meter-reading capture from device camera, cutting manual data entry errors by 90%.
- Developed a Flutter enterprise application (Bloc) for a factory survey system with full API auth, permissions, media upload, and step-by-step data submission flow.
- Built an iOS Voice Assistant SDK (Swift, UIKit, XCFrameworks) enabling real-time voice-driven banking transactions via full-duplex WebSocket audio streaming.
Android Developer
Promoted from Intern — Sep 2022
ISSM — Islamabad, Pakistan
- Refactored legacy Android codebase into MVVM Clean Architecture, enabling a maintainable and testable codebase for critical banking features.
- Built a contactless biometric authentication feature for an enterprise banking app with 99% acceptance accuracy.
- Developed a voice assistant for a fintech banking app, reducing multi-step flows to 2-step voice interactions and improving task completion speed by 60%.
- Built a real-time full-duplex audio streaming module using WebSockets with on-device voice activity detection and under 100ms latency.
- Designed a custom CameraX + TFLite pipeline processing live frames with under 200ms response time.
- Achieved approximately 40% reduction in battery drain through efficient thread and model lifecycle management during AI inference sessions.