Week 10

Feature: Recommendation Algorithm Design

A major focus this week was developing the foundation of our recommendation algorithm.

Rather than jumping straight into code, we worked on:

• Defining the balance between relevance and randomness to prevent repetitive content

• Exploring how to incorporate filters and user preferences without overcomplicating the experience

• Structuring key inputs such as:

  • • AI-optimized keywords

    • Duration and content filters

    • Avoided words and themes

• Planning for future scalability, including personalization based on user behavior

This approach ensures that when we implement the algorithm, it is both intentional and adaptable.

User Feedback & Validation

We also spent time speaking with potential users to better understand how AuroraTV should feel in practice.

Key takeaways:

• Users prefer quick and effortless discovery over highly precise recommendations

• A simple interface is more appealing than one with too many controls

• The “TV-like” passive viewing experience is a strong differentiator

• Smooth content flow matters more than perfectly curated results

These insights are directly influencing both our algorithm design and UI decisions.

Team Building & Brainstorming

This week also emphasized collaboration and alignment within the team. We:

• Brainstormed ideas for upcoming features such as the Friends Center and social sharing

• Discussed long-term goals and product direction

• Refined responsibilities across frontend, backend, and AI integration

• Explored ways to keep the app intuitive while still feeling intelligent

Looking Ahead

Next week, we plan to:

• Begin implementing the refined recommendation algorithm

• Test early versions with real use cases

• Continue iterating on UI based on feedback

• Progress toward personalization features