How We Are Building a Chatbot That Helps You Design Your Own Jewelry

Published On: 9 October 2025.By suman yubraj.

How We Are Building a Chatbot That Helps You Design Your Own Jewelry

Angara AI Ring Assistant | Case Study | Auriga IT

Case Study — Retail AI

Angara AI Ring
Assistant

A Personal Jeweler, Available 24/7

How Auriga IT built a conversational AI that turns love stories into bespoke ring recommendations — bridging the emotional gap between a customer's feelings and Angara's fine jewellery catalogue.

About Angara

Founded in 2005 by Ankur and Aditi Daga, backed by a 300-year gemstone legacy, Angara is a D2C, vertically integrated fine jeweller that controls everything from sourcing gemstones to crafting final pieces. The brand's online customization tools are among the most robust in the industry — yet a gap remained: customers still missed the one-to-one guidance of an in-store expert.

Bridging that gap — at scale, 24 hours a day — became the brief for Auriga IT.

The Challenge: Too Many Choices, Not Enough Guidance

Imagine searching for the most important ring of your life — but feeling like you need to be a gemologist to find it. Many of Angara's customers arrived knowing the feeling they wanted ("timeless," "adventurous," "delicate") but had no way to translate that feeling into a product search. The gap between emotion and a filter-based catalogue created three compounding problems:

Confusion — Thousands of ring options with no personalised entry point left customers feeling overwhelmed rather than excited.
Frustration — Mapping a personal vision to technical website filters (cut, carat, setting style) was counterintuitive and friction-heavy.
Doubt — Without expert guidance, customers feared making an expensive mistake on a high-emotion, high-value purchase.

The Solution: Story-Driven AI Personalisation

Auriga IT built an AI ring design assistant that changes the conversation entirely — starting not with a catalogue, but with a love story. The journey unfolds in five intelligent steps.

It Starts With Your Love Story

Instead of a generic catalogue, the AI opens with a simple invitation: "Tell me your story." Customers chat naturally — sharing how they met, what makes their partner unique, what feelings define their relationship.

It Creates a Ring Personality Profile

As the conversation unfolds, the AI identifies emotional themes — strength, elegance, love for nature, adventure — and builds a unique design profile for the kind of ring that would feel like a perfect fit.

You Swipe Through a Personal Catalogue

The AI presents a curated ring collection. Customers swipe right on designs they love, left on designs they don't — giving the AI instant visual feedback on style preferences without additional conversation.

It Becomes Your Design Partner

The AI takes "liked" rings and explains precisely why each one matches the customer's story. For example: "Since you mentioned your ten-year journey together, I suggest this durable platinum band — representing strength and lasting love."

Live Customisation on Demand

Found a ring they almost love? Customers say: "Show me that in rose gold" or "Can I see a slightly bigger diamond?" — and the AI updates the design instantly, making the customer an active co-creator.

90-Second Micro-Journey

Angara AI Ring Assistant — Live Demo

0 – 5 seconds · Welcome

Welcome to Angara! I'm your personal ring design assistant. I'm here to help you create a truly special ring that tells your unique love story. Ready to begin? ✨

👤

Yes, absolutely! We're looking for an engagement ring.

5 – 25 seconds · Sharing the Love Story

Wonderful! To start, tell me a little about your love story. How did you two meet, or what's a special memory you share?

👤

We met hiking in the Dolomites. It was a complete accident, but we ended up spending the whole trip together — laughing and exploring. She's so adventurous and loves nature.

25 – 45 seconds · AI Creates Profile & Curated Swipes

That's a beautiful story! Adventurous and love for nature — I've noted those themes. I've curated a few ring styles for you. Swipe right on what catches your eye, left on what doesn't. No pressure — just go with your gut! 💍

Technology Architecture

Every layer of the stack was chosen to serve a specific engineering need — from natural language understanding through to observable, production-safe deployment.

Language Understanding

OpenAI LLM

Parses plain, emotional customer language into structured design choices and preference signals the system can act on.

Conversation Orchestration

LangGraph

Manages multi-step, stateful conversations — remembering preferences across turns and guiding customers to a confident finish.

Quality & Observability

Langfuse

Traces every interaction for speed, safety, accuracy, and cost control — essential for responsible beta deployment and continuous improvement.

Runtime & APIs

Python · FastAPI

Production-ready session management, API endpoints, and downstream integrations with Angara's catalogue and customisation systems.

RingDNA — Proprietary Mapping Layer

The secret ingredient. RingDNA is Auriga IT's internal mapping engine that translates emotional language and lifestyle cues into precise, actionable ring attributes — bridging the gap between what a customer feels and what the catalogue contains.

vintage delicate ocean-blue low-profile → halo vs. bezel prong count head height shank profile metal palette

The Impact

Emotion-to-Product Bridge

Customers can now find their perfect ring using feelings and stories — not gemological vocabulary or complex filters.

24/7 Expert Availability

An in-store-quality guided experience is now available at any hour, eliminating the dependency on live sales staff for high-intent customers.

Reduced Decision Anxiety

Story-driven personalisation and contextual recommendations directly address doubt — the single largest barrier to conversion on high-value purchases.

Live Co-Creation

Real-time customisation requests transform passive browsing into active design engagement, increasing purchase commitment and reducing returns.

Observable & Improving

Langfuse observability ensures every interaction is measured — enabling continuous improvements to accuracy, speed, and cost-efficiency post-launch.

Reusable AI Architecture

The RingDNA mapping layer and LangGraph orchestration pattern are reusable across other high-consideration D2C product categories beyond jewellery.

Build AI That Understands Your Customers

Does your product require personalisation that goes beyond filters and keywords? Auriga IT specialises in conversational AI, agentic systems, and LLM-powered experiences that turn customer intent into business outcomes.

Talk to Auriga IT