Klyq AI
Klyq AI is an agentic marketing platform for e-commerce brands. AI agents monitor store and ad data, generate and test campaigns, and push suggested actions to a dashboard or Slack where the user approves or rejects them.
The Problem
E-commerce brands drown in scattered store and ad data and spend hours manually monitoring performance and running campaigns. Klyq turns that data into action, with agents doing the monitoring and campaign work and a human staying in control via approve/reject.
What I Built
As full-stack and AI engineer, I built the frontend, the backend, and the agentic flow.
- The frontend dashboard where users see insights and approve or reject actions
- The backend services that ingest store and ad data and power the platform
- The agentic flow: AI agents that monitor data, generate and test campaigns, surface insights, and push suggested actions for the user to approve
- A plain-English chat so users get answers without touching a dashboard, over a unified store/ads/inventory data model that learns with every sale
The Hard Part
Getting the agents to produce reliable, useful actions from messy multi-source data rather than noise, and designing the human approval loop so nothing runs without sign-off.
Tech Stack
- Node.js
- OpenAI
- Gemini
- LLMs
