For coffee lovers, the pursuit of the perfect cup can feel like an endless quest. With a countless array of beans, roasts, and flavor profiles to choose from, finding the one can be overwhelming. But what if there was a way to simplify the search? Enter Coffee Nirvana, an AI-powered solution that harnesses the power of Oracle on Google Cloud that was demonstrated at Oracle CloudWorld 2024 this week. With Coffee Nirvana, no more endless scrolling through online stores or disappointing cafe orders. Simply input your preferred taste and strength, and let Coffee Nirvana do the rest. 

Challenges of traditional search

For certain kinds of hunts, traditional search methods can be a real buzz kill, often relying on keyword matching, limiting the accuracy and relevance of results, particularly when dealing with nuanced descriptions like "a smooth, medium-roast coffee with hints of chocolate and hazelnut." To help, Coffee Nirvana uses Retrieval Augmented Generation (RAG), which empowers AI to tap into vast databases of information that provide real-time, context-specific results. In the world of coffee, this means a search experience that's as nuanced as your taste buds.

How RAG powers Coffee Search

1. Oracle Database@Google Cloud: Oracle Database version 23ai deployed on Google Cloud.
2. Vector embeddings: Coffee descriptions and user queries are transformed into numerical representations (vectors) using a Vertex AI large language model.
3. Semantic search: Langchain facilitates a similarity search within the vector database (populated with sample coffee data in the Oracle database), retrieving the most relevant matches to the user's query.
4. Contextual understanding: The Gemini Flash model adds another layer of sophistication, refining the search based on contextual cues and nuances in the user's description.
5. Google Maps: The Google Maps API adds yet another layer of visualization, displaying the location of stockists of the coffee.

Databases and embeddings

The solution also relies on a database that stores vectors, a crucial part of storing and efficiently retrieving embeddings. Google Cloud offers tools like Vertex AI Vector Search, and in this demo, we leverage the power of an Oracle database to store and manage these embeddings effectively.

 See it in action

The power of Oracle and Google Cloud

The Coffee Nirvana demo showcases a number of benefits of using RAG with Oracle Database on Google Cloud:

• Precision and personalization: No more sifting through irrelevant results. RAG ensures that the recommended coffees genuinely align with your unique tastes.

• Real-time inventory: Integrate with your Oracle database to provide up-to-the-minute availability information, enhancing the user experience.

• Natural language understanding: Express your coffee desires in your own words, fostering a more intuitive and engaging search process.

By combining the power of LLMs with real-time, context-rich data stored in an Oracle database, this approach delivers personalized recommendations, elevating the user experience. The potential applications of this technology extend far beyond the realm of coffee. From retail and hospitality to healthcare and education, this AI-powered solution has the potential to revolutionize industries by providing tailored experiences that cater to individual needs and desires. The fusion of AI and data is not just the future of search — it's the future of personalized discovery. But don’t just take it from us. Here’s what Dun & Bradstreet, a global leader in business decisioning data and analytics, had to say.

“With the unparalleled quality of D&B data, we can now seamlessly integrate Oracle Database’s performance, reliability, and scalability with Google Cloud’s powerful analytics and AI tools. This synergy allows us to process and analyze massive datasets with unprecedented speed and efficiency, extract deeper insights, and deliver more value to our customers.” - Adam Fayne, VP, Enterprise Engineering, Dun & Bradstreet

Ready to get started with Oracle Database on Google Cloud for your next-generation AI-powered applications? Check out the following resources:

• Stay tuned: Follow us for updates on future developments and integrations.
• Explore further: Learn more about RAG and vector databases on Google Cloud.
• Get started: Oracle on Google Cloud
• Explore the code: Github