Architected a multi-agent AI system that automates paid-advertising campaigns end-to-end. Designed the digital-twin model that simulates campaign performance before spend is committed, and built the full web platform operators use to launch and monitor ads.
I design and ship
multi-agent AI systems
that run in production.
Builder first. Agent engineer by trade.
I build production multi-agent systems on the Claude Developer Platform — and the full-stack web platforms that run them. From architecture and digital-twin design to deployment, I take ideas from the first call to a system that ships and stays up.
Most of my work lives at the seam between AI engineering and product. The bottleneck on a working agent product is rarely the model — it's whether someone in the room can translate "we want an agent to do X" into a concrete architecture, an eval set, and a deployment plan. That's the part I like.
I work with founders directly. I scope, pick the stack, design the evals, and own the build through release.
Three platforms, three problems, all in production.
Top Rated profile delivering production multi-agent architectures, workflow automation, and voice-AI systems for clients in advertising, sales, education, and government communications. Own scoping, architecture, build, and deployment — from first call to running system.
Shipped a full booking and operations platform that lets local businesses run their day from one dashboard. Built an internal wallet handling online payments, balance tracking, and withdrawals, plus the multi-agent automations that take repetitive admin work off the owner's plate.
Tools I reach for first.
- Python
- TypeScript
- Node.js
- Claude Agent SDK
- Multi-agent orchestration
- Prompt engineering
- RAG
- Voice AI
- Full-stack web apps
- Payments & wallets
- Workflow automation
- Production deployment