SophieMBerger is a mobile-focused developer with a strong emphasis on iOS (Swift) and emerging skills in Machine Learning and Python scripting. Her portfolio consists largely of educational coursework and hackathon projects, demonstrating a proactive learning approach but showing typical early-career patterns in code maintenance and architecture. She has successfully integrated ML models into mobile environments, though her tooling often relies on legacy dependencies.
Score Context: Score reflects current portfolio maturity rather than potential. Fundamentals in mobile and scripting are present, but the repository practices (hardcoded paths, legacy dependencies) indicate a developer still transitioning from guided learning to professional production standards.
🏞A TensorFlow Lite implementation of Google NIMA (Neural Image Assessment)
An easy way to keep track of your vaccinations 💉(2nd place at Microsoft 6 hour hackathon)
📁 I decided to create a Python script that automatically moves this folder into a folder I created just for these type of files within my Documents directory.
😸🐑This is an image classifier which was trained to distinguish between images of my cat "Minnie" and sheep. It makes sense once you see my cat!
Repositories frequently contain committed dependency folders (Pods/) and hardcoded local user paths (e.g., /Users/SophieMBerger/...), indicating a need for better .gitignore usage.
High rate of finishing projects, including UI polish and functional logic, particularly in her mobile coursework and hackathon entries.
Shows initiative in combining domains, such as running Neural Image Assessment on mobile devices and participating in hackathons.
Demonstrated ability to build functional apps with networking (Alamofire), UI (Auto Layout), and delegates, though largely within the context of structured coursework.
Writes functional automation scripts (MoveRelocatedItems), but code analysis reveals brittleness: hardcoded absolute paths, lack of configuration, and minimal error handling.
Experience with Image Classification and Sentiment Analysis; however, projects rely on obsolete TensorFlow 1.x syntax and lack clean dataset management.
Understands MVC patterns, protocols, and delegates as evidenced by her iOS repo descriptions, but lacks exposure to modern architectures like MVVM or TCA.
Can implement scrapers (LinkedIn Parser) using Selenium/BeautifulSoup, though they lack production features like explicit waits or headless execution.
Get docs, diagrams, scorecards, and reviews for any repository. Understand code faster.