Answer from real business data
AI uses your services, prices, policies, and availability to answer customers on LINE OA with the right context.
AI BUSINESS PLATFORM · AVAILABLE NOW
AI that turns LINE conversations
into completed work.
iLineme connects conversations, business knowledge, availability, pricing, deposits, and back-office operations in one flow. Customers get faster answers while your team keeps visibility and control over every important task.
Visit iLineme ↗AVAILABLE NOW
LINE OAAI ASSISTANTBOOKING + PAYMENTTEAM OPERATIONS
01 / PRODUCT
One conversation.
One operating flow.
Designed for service businesses, lean teams, and operators who want to serve customers on LINE without duplicating work behind the scenes.
Move from chat to booking
Check availability, select a service, collect customer details, calculate pricing, and create a booking without switching systems.
Bring in the team at the right time
Give operators the conversation, status, and information already collected by AI so customers never need to start over.
02 / DEVELOPMENT
Built for the moment
chat becomes business.
Chat requires real-time event handling, but booking, payment, team permissions, and tenant data still need transactional correctness and traceability.
The architecture separates customer messaging from asynchronous AI work while PostgreSQL remains the transactional source of truth. Small-model orchestration combines targeted retrieval, compact context, and tools to keep responses fast and cost-efficient without giving the model responsibility for transactional state.
Next.js application
Next.js, React, and TypeScript connect the customer experience, LIFF, and operations console in a codebase with clear boundaries.
Multi-tenant data model
PostgreSQL and Prisma isolate tenants, projects, members, customers, conversations, services, and bookings with authorization at every layer.
Async message pipeline
LINE webhook signatures are verified before work enters Google Cloud Pub/Sub, separating fast acknowledgements from longer AI workflows.
Workflow + RAG
A workflow engine coordinates intent, retrieval, tool execution, and response using embedding caches, business knowledge, and release snapshots.
Small-model LLM orchestration
Compact context, targeted retrieval, structured outputs, and tool-based task decomposition let smaller LLMs handle focused work accurately while reducing latency and cost per conversation.
Reliable commerce flows
Booking and payment paths use idempotency, conflict checks, transaction boundaries, and auditable state transitions.
Operational controls
Redis supports rate limiting and caching, Google Cloud services handle events and artifacts, and draft-to-release workflows protect live customer traffic.
MESSAGE EXECUTION MODELVERIFIED · ASYNC · TENANT-AWARE
LINE WebhookSignature verify · tenant resolve · fast acknowledge
→Pub/Sub + AI FlowIntent · retrieval · tools · guardrails
→Customer + OpsLINE response · booking · handoff · audit trail
AVAILABLE NOW
Make LINE work
like your best operator.
Explore iLineme now, or talk to Hearro if you need an AI, booking, or operational workflow tailored to your business.