ChatC AI Chatbot for WooCommerce

Role

Backend and plugin integration engineer.

Business Problem

E-commerce stores needed automated customer support that could answer product-specific questions accurately. Generic chatbots hallucinated product details. The solution needed to stay current as products were added/removed without manual retraining.

Key Decisions

• Implemented a RAG pipeline that automatically re-embeds product data on WooCommerce webhook events (product created/updated/deleted).
• Used Qdrant vector database for semantic search over product embeddings, with fallback to keyword search.
• Built the WordPress plugin with a floating widget that required zero frontend framework dependencies.

Architecture

• Built .NET API integrations with OpenAI-compatible provider support and Qdrant vector storage.
• Created WordPress plugin UX with admin settings and floating chat widget.
• Implemented automated product embedding pipeline triggered by WooCommerce webhooks for real-time catalog freshness.

Outcomes

• Enabled e-commerce teams to deploy smarter support automation with low integration friction.
• Improved response relevance by grounding assistant answers in current product data via RAG.

Measurable Impact

• Automated product embedding pipeline with webhook-triggered updates
• Zero-dependency floating chat widget for WordPress
• Support for multiple OpenAI-compatible LLM providers