On the DL
ReactNext.jsTypeScriptTailwind CSS

On the DL is a web application developed in March 2025 as part of a consulting engagement, built for two well-traveled friends who are foodies and socialites. They had been sharing recommendations with friends — from party venues to food spots and more — but had been storing all their information in a Google Doc.

The platform transforms their curated recommendations into an interactive AI-powered experience. Users can select which founder they’d like to chat with, and then engage with an AI that’s trained on that founder’s recommendations. The AI provides personalized recommendations based on the user’s specific needs and preferences, using Retrieval Augmented Generation (RAG) to ensure accurate, contextually relevant suggestions.

⸻

The Vision

The vision behind On the DL is to make curated, personal recommendations accessible and interactive. Instead of scrolling through a static document, users can have a conversation with an AI that understands the founder’s taste, preferences, and expertise.

We believe that when recommendations are personalized and conversational:

  • Users get exactly what they’re looking for through natural dialogue
  • The knowledge and expertise of well-traveled insiders becomes accessible
  • Recommendations feel more personal and trustworthy
  • The discovery process becomes engaging and interactive

Great recommendations shouldn’t be buried in documents. On the DL brings them to life through conversation.

⸻

The Platform

On the DL provides an intelligent recommendation platform:

  • Founder Selection: Users can choose which founder they want to chat with based on their expertise
  • AI-Powered Chat: Conversational interface with AI trained on each founder’s recommendations
  • Personalized Recommendations: Get tailored suggestions for parties, food, venues, and more based on your specific needs
  • Retrieval Augmented Generation: Uses RAG to provide accurate, contextually relevant recommendations from the founder’s knowledge base
  • Natural Conversation: Ask questions, specify preferences, and get recommendations through a natural chat interface

⸻

Technologies Used

  • Web Application

    • Next.js for server-side rendering and optimal performance
    • React for component-based UI architecture
    • TypeScript for type safety and developer experience
    • Tailwind CSS for responsive, utility-first styling

  • AI & Recommendations

    • Retrieval Augmented Generation (RAG) for contextually accurate recommendations
    • AI model training on founder-specific recommendation data
    • Natural language processing for conversational interactions
    • Knowledge base integration from Google Docs and curated recommendations

  • User Experience

    • Interactive chat interface
    • Founder selection and profile system
    • Personalized recommendation engine
    • Real-time conversation flow