Case Study · Social & Entertainment
SeLeBox
SeLeBox began as an entertainment platform centered around audiobook and story content. As a Full-Stack Engineer at YenkoDev, my work focused on extending the existing production application with interconnected social features - including profile social graphs, direct messaging, group chat, user-generated media feeds, continuous vertical video discovery, and offline media availability.
Overview & Product Narrative
SeLeBox is a web application originally built to deliver digital audiobook and story content to users. To increase platform engagement and transform passive content consumption into an active community experience, the platform required a comprehensive social and media feature expansion.
Rather than replacing the established product, the engineering objective was to extend the live application with interconnected communication tools, user-generated media publishing, and modern video discovery patterns without interrupting active user accounts or existing monetization features.
Existing Entertainment Platform
↓
Audiobooks & Story Content
↓
Need for Richer User Interaction
↓
Social & Media Feature Expansion
├── Follow / Follow Back Graph
├── Direct & Group Messaging
├── User Media Feeds & Uploads
├── Continuous Vertical Video
└── Offline Content Caching
↓
Interactive Social Entertainment Hub
Existing Platform Context
Before the social expansion, SeLeBox operated as a digital content platform. Established platform capabilities included:
Note: These core capabilities establish product context and pre-existed my major social feature contributions.
My Role & Contribution Boundary
As a Part-Time Full-Stack Engineer at YenkoDev, my responsibilities centered on implementing frontend UI/UX components, integrating backend application services, developing social and media features, and maintaining existing codebase areas.
EXISTING PLATFORM
- Audiobook streaming engine
- Romantic story catalog
- Virtual coin economy
- Coin purchasing workflows
- Base user account schemas
- Initial responsive UI templates
MY CONTRIBUTIONS
- Follow / Follow Back social relationship graph
- Direct Chat person-to-person messaging
- Group Chat multi-user conversations
- Feed services & post retrieval
- Image and video upload pipeline
- Continuous vertical video feed
- Like, Comment, and Share interactions
- Offline video caching implementation
- Appwrite service integration & maintenance
Product Evolution Breakdown
The addition of social capabilities transformed SeLeBox from a one-way content consumption app into a two-way social entertainment ecosystem. Users who previously visited solely to listen to audiobooks could now follow creators, join discussion group chats, post user-generated media, and consume vertical video clips.
Interconnected Feature Architecture
Rather than existing as isolated modules, new features were engineered as an interconnected workflow where profile relationships directly empowered communication and media sharing:
User Profiles
↓
Follow / Follow Back Graph
↓
┌──────────────────────┬──────────────────────┐
│ │ │
Messaging Subsystem Feed Subsystem Social Discovery
│ │ │
├── Direct Chat ├── Media Post Uploads ├── Continuous Vertical Video
└── Group Chat └── Feed Retrieval └── Like / Comment / Share
↓
Media Consumption
↓
Offline Video Caching
Messaging Subsystem - Direct & Group Chat
Designed and integrated communication tools to foster user interaction directly within the application.
User A Initiates Chat
↓
Check Relationship with User B
↓
Valid Social Connection?
┌─────┴────────────────┐
NO YES
↓ ↓
Prompt Follow Load / Create Conversation
↓
Send / Receive Messages
Feed Services & Media Uploads
Developed user-generated feed services allowing community members to publish visual updates, story reviews, and media clips.
The media upload pipeline handles post composition, file validation, storage upload, metadata persistence, and feed list updates.
User Creates Post
↓
Attach Image / Video File
↓
Upload File to Appwrite Storage
↓
Persist Post Metadata (Caption, Author, File Ref)
↓
Update Feed List Output
Continuous Vertical Video Experience
Implemented a continuous vertical video browsing feed providing an infinite content stream optimized for mobile and desktop screens.
Features and interactions:
- Continuous Video Playback: Smooth auto-play transition as the user scrolls vertically through video clips.
- Like Interactions: Instant feedback toggle allowing users to appreciate video content.
- Comment Threads: Contextual comment overlays for community discussion.
- Social Sharing: Action triggers enabling content link propagation.
Offline Video Caching
Implemented client-side caching mechanisms for selected video content to support playback under limited or interrupted network connectivity.
When users mark videos or access content under stable connections, media resources are cached locally so playback remains seamless during network drops.
Backend Services - Appwrite & Firebase Transition
Development was executed in a dual backend-service environment. The engineering direction focused on gradually shifting application services toward Appwrite while maintaining legacy integrations with Firebase.
Worked within a backend-service environment using Appwrite alongside remaining Firebase functionality, with the application gradually moving away from Firebase.
React Frontend Application
↓
Application Integration Layer
↓
Appwrite Services (Primary Growth)
├── Authentication & Sessions
├── Database Documents & Feeds
└── File Storage for Media Uploads
↓
Firebase Services (Diminishing Legacy Role)
System & Production Constraints
Key Technical Decisions
1. Extend Existing Platform Architecture
Context: Production platform had live users and established coin monetization workflows.
Decision: Extend existing React frontend components and integrate new Appwrite services without disrupting core audiobook behavior.
2. Couple Direct Messaging to Social Relationship Verification
Context: Unrestricted user messaging risked spam and poor user experiences.
Decision: Require verified Follow/Follow Back status before initiating direct chat conversations.
3. Streamlined Media Upload Pipeline
Context: User-generated feeds required reliable image and video file handling.
Decision: Direct file uploads to Appwrite Storage with atomic document creation for post metadata.
4. Continuous Vertical Video Stream Integration
Context: Platform needed high-engagement video discovery functionality.
Decision: Build a continuous vertical video feed combining auto-play behavior with contextual Like, Comment, and Share interactions.
5. Client-Side Video Caching
Context: Mobile users experienced buffering interruptions on unstable networks.
Decision: Implement video resource caching for selected content to preserve playback continuity offline.
Verified Qualitative Outcomes
Lessons & Retrospective
Interconnected Product Engineering
SeLeBox reinforced that extending an existing product requires understanding how seemingly separate features affect one another. Profiles, follow graphs, direct chat, group rooms, media uploads, and video feeds became interconnected product behaviors rather than independent pages.
Evolving Service Boundaries
Working within an environment shifting from Firebase to Appwrite emphasized the importance of writing adaptable frontend integration code that abstracts backend service specifics behind clean data handlers.
Social Relationships - Follow / Follow Back
Implemented profile-level social relationship capabilities. The social graph enabled users to connect with story creators and fellow community members.
Key functionality: