Mohamed Aziz Ben Salem

• Owned architecture-level decisions across 2 cross-platform Flutter apps — designed and implemented
Clean Architecture with BLoC state management, enforcing OOP/SOLID principles across the codebase
for scalability and maintainability.
• Reduced app crash rate by 60% through rigorous testing protocols, performance profiling, and
systematic debugging across iOS and Android builds.
• Built and maintained automated CI/CD pipelines using Fastlane and Codemagic — doubled release
frequency (monthly → bi-weekly) with zero undetected regressions.
• Implemented Dart Streams for real-time data flows and async state management, improving perceived
app responsiveness by 20% through local persistence with Sqflite and Hive.
• Cut new feature development time by 15% by architecting a modular, Provider-based state management
system with clear separation of concerns.
• Mentored junior developers through code reviews and pair programming sessions, enforcing best
practices in Dart, architecture, and testing.

• Delivered a fully responsive, production-ready cross-platform Flutter app — pixel-perfect adherence to Figma
mockups across iOS and Android device configurations.
• Architected the app using MVVM pattern with BLoC/Provider state management, cleanly decoupling business
logic from UI for testability and long-term maintainability.
• Built complex custom widgets and animations using Flutters rendering pipeline, significantly improving user
engagement and visual polish.
• Integrated RESTful APIs and implemented WebSocket connections for real-time data synchronization,
handling all async flows with Dart Streams.

• Coached 20+ junior and mid-level developers in Flutter, Dart, mobile architecture patterns, and software testing
— achieving a 4.7/5 satisfaction rating.
• Redesigned curriculum to cover architecture principles (MVVM, Clean Architecture), OOP/SOLID, state
management, and real-world CI/CD workflows.
• Led students through end-to-end project delivery: from architecture design to production release on Google
Play and the App Store.

• Built and maintained the Sweaka Supervisor app — a cross-platform mobile application serving 10, 000+ active
users for authentication, data retrieval, and transaction processing via RESTful APIs.
• Reduced crash reports by 20% while delivering the most visually praised UI update per post-release user
surveys.

Developed a Named Entity Recognition (NER) pipeline to automatically extract structured entities, such as people, organizations, locations, and dates, from unstructured news articles.

Built and fine-tuned NLP models using spaCy, improving extraction accuracy through custom training on domain-specific datasets.
Preprocessed and cleaned large volumes of raw text data to prepare training corpora for model development.

Collaborated with the engineering team to integrate the entity extraction module into the company's existing data pipeline.

Contributed to the development of a marketplace web application built with Angular, implementing reusable UI components and dynamic product listing features.

Integrated RESTful APIs to handle product data, user interactions, and real-time updates across the marketplace interface.

Collaborated with the team to deliver responsive, cross-browser compatible front-end features aligned with UX requirements.