Rabbit AI | Search & Discovery

Tubi

Tubi's search struggled when viewers didn't know what to watch. Users needed a more intuitive way to discover content through moods, preferences, and natural language.

The core idea: What if we replaced keyword based search with descriptive search?

I helped shape Rabbit AI from concept through iteration. I advocated for embedding conversational AI into search rather than relying on a standalone chatbot.

Rabbit AI drove 18% more search refinements and 12% higher click-through rates. Subsequent iterations reduced measured churn from 56% to 39%, improving content discovery and engagement.

Employer:

Me + PM + DS + DA + iOS Engineers + QA

Role:

Lead Product Designer

Year:

2024

Audience:

1 Million+ iOS users

Areas of influence:

Highlights:

Explore the full story –

Challenge

At Tubi, I worked on Rabbit AI, an AI-powered experience designed to help viewers discover what to watch. Traditional streaming search worked well when users knew exactly what they wanted, but struggled when they searched using moods, preferences, or vague descriptions. Our challenge was to make discovery more conversational and intuitive, helping users find relevant content without knowing specific titles or genres.

Objective

I partnered with product, engineering, and AI teams to explore how large language models could improve content discovery. We initially explored a standalone conversational assistant, but I saw an opportunity to integrate AI more directly into the existing search experience. My focus was on three principles: understand natural-language intent, make recommendations relevant and explainable, and help viewers move from exploration to watching with minimal friction.

Results

I helped shape the experience from early concepts through experimentation and iteration. First, we launched a conversational AI experience to validate whether viewers would engage with natural-language discovery. Second, I advocated for bringing AI capabilities closer to search, reducing the friction of switching between a chatbot and the core browsing experience. Third, we refined recommendation presentation and conversational interactions based on how users searched, explored, and selected content. A key product tradeoff was balancing the novelty of a standalone AI assistant against the usability and discoverability of integrating AI into familiar workflows.