I work with founders and product teams on AI systems that run in production. Every engagement starts with a short discovery call so scope, timeline, and price can be agreed in writing before any work begins. Prices below are indicative starting points in USD and are finalised in a written proposal.
Production AI agents that actually run in production — voice, chat, and workflow. From Vapi receptionists that book appointments and dispatch jobs, to WhatsApp / web-chat agents with per-tenant configuration, to multi-step LangGraph workflows that hand off to humans when they should.
Python/FastAPI backends, API integrations, and internal tools. For teams that need a senior engineer to ship a well-scoped system without hiring full-time.
End-to-end business process automation with n8n, Make.com, and custom code — the kind that runs quietly and saves a team from hours of manual work each week.
Retrieval, tool use, and agentic workflows built on Claude, OpenAI, and the Model Context Protocol. Includes MCP server design for teams exposing internal systems to LLMs.
A focused review of an existing AI, backend, or automation system, with a written report of what to fix, what to keep, and what to build next.
How pricing works
Every project is quoted in a written proposal after a discovery call. Fixed-scope projects use a 30–50% deposit and milestone payments. Retainers and hourly work are billed weekly or monthly in arrears.
Payment terms, deliverables, and cancellation rights are governed by the signed proposal, the Terms of Service, and the Refund & Cancellation Policy.