← ALL WORKCASE STUDY 006 · AI

WAGRAICHAT

TOTO — AI PET ASSISTANT
CLIENT

WAGR

YEAR

2026

ROLE

AI product design · RAG pipeline · Chat UI · Background workers

STACK

Next.js · NestJS · OpenAI · Vector embeddings · Supabase · BullMQ

VISIT LIVE SITE
WAGR AI CHAT
[OVERVIEW]

[THE CHALLENGE]

Generic chatbots give risky pet advice and can't recommend anything real. WAGR needed an assistant grounded in its own knowledge and catalogue.

[THE SOLUTION]

A retrieval-augmented assistant built on WAGR's knowledge base, with a findings panel that turns answers into actions — book a groomer, open a pet profile or shop the right product.

[WHAT I BUILT]
24/7INSTANT ANSWERS
3ACTION TYPES FROM CHAT
1GROUNDED KNOWLEDGE BASE
NEXT CASE STUDYDEAL COLLAB AI