Beeflection
Turning project experiences into career insights
Beeflection is a reflection app designed to help users identify both soft and hard skills derived from their project experiences.

The Journey
Identifying Hidden Skills
The problem with self-assessment
As developers, we often struggle to articulate our skills. We know we've built projects, solved problems, and collaborated with teams - but when it's time to update our resume or prepare for an interview, we blank out. The problem: • Hard to objectively assess your own growth • Skills learned in projects aren't always obvious • No structured way to track professional development • Difficult to categorize skills by job role (Tech, Design, Product) I wanted to build an app that would help people recognize and document the skills they're actually developing through their work.
AI-Powered Skill Extraction
Using Gemini to analyze reflections
The core idea: users write reflections about their project experiences, and AI extracts the skills they demonstrated. For example, if someone writes: "I led the team through a challenging deadline, coordinated with designers, and refactored the codebase for better performance" The AI would identify: • Leadership (soft skill) • Communication (soft skill) • Code Optimization (hard skill - Tech) • Project Management (soft skill) I integrated Google's Gemini AI for this. The data flow: 1. User completes a reflection 2. App sends reflection text to Gemini 3. Gemini analyzes and categorizes skills 4. Skills are saved to SwiftData with categorization The MVVM architecture kept business logic (AI integration) separate from UI, making it easy to test and iterate.
Building Under Pressure
One week from concept to completion
We had one week to build this app as part of Apple Developer Academy. I was responsible for the iOS development while working with designers on the UI/UX. Day-by-day breakdown: • Day 1-2: SRS, ERD, UML diagrams (planning is crucial) • Day 3-4: Core functionality (SwiftData models, Gemini integration) • Day 5-6: UI implementation, polishing • Day 7: Testing, bug fixes, final touches The biggest technical challenge? SwiftData relationships. I needed to model: • One reflection → many skills • Skills categorized by role (Tech, Design, Product) • Filtering and searching reflections The solution was careful schema design upfront, with clear relationships defined before writing any code.
Career Growth Tool
Helping developers see their progress
After launch, users started creating reflections regularly. The pattern was clear: when people actively reflected on their work, they became more aware of their skills. One user said: "I never realized I'd developed so many soft skills until Beeflection showed them to me. It made updating my resume so much easier." The app served two purposes: • Self-awareness: Recognize skills you're developing • Documentation: Build a portfolio of your growth For me, this project proved that rapid development doesn't mean sacrificing quality. With good planning (SRS, ERD, UML) and clear architecture (MVVM), you can deliver polished work in tight timeframes.
Technologies Used
Key Features
Key Learnings
Planning saves time - the SRS, ERD, and UML work upfront made development smooth
SwiftData relationships need careful design - you can't easily change schema later
AI integration is powerful but needs good prompts - garbage in, garbage out
MVVM architecture is worth the overhead - it made the codebase maintainable even under time pressure
User testing is crucial - our initial UI was too complex until we simplified based on feedback
Project Gallery



Interested in this project?
Feel free to explore the code or see it in action. I m always happy to discuss the technical details!