A/B Test & Feature-Flag MCP
MCP servers that let AI agents safely read and manage feature flags and experiments — rollouts and A/B tests an agent can reason about.
Hi, I'm
AI & Full-Stack Engineer
I build intelligent products — from AI agents, MCP and prompt systems to fast, well-architected web apps. Based in Ukraine 🇺🇦
I'm a Full-Stack & AI Engineer who loves turning fuzzy ideas into clean, shippable products. I work across the stack — Python, TypeScript, React, Next.js and Nest.js — and I'm deep into the AI space: building with the Model Context Protocol, the Claude Agent SDK, prompt engineering and automation.
I care about good architecture, fast feedback loops and finding the real problem before writing a line of code. I also run tech interviews and enjoy helping engineers level up.
Away from the keyboard you'll find me on a tennis court or a yoga mat — and often coaching both. Sport taught me the discipline, patience and calm I bring to engineering.
A toolkit spanning AI engineering, full-stack development, cloud and the craft around shipping software.
Day to day I build AI-agent infrastructure on Claude and the Model Context Protocol. A high-level look at the kind of things I ship.
MCP servers that let AI agents safely read and manage feature flags and experiments — rollouts and A/B tests an agent can reason about.
An MCP server that exposes a content management system to Claude, so structured content can be created and edited through natural language.
Authentication and scoped access control for MCP servers and agents, so they only ever act with the permissions they should have.
Tools, skills and automations on the Claude Agent SDK — plus the prompt systems that keep agents reliable and repeatable.
The parts of me that don't fit in a commit. Swipe through 👇
A few favourite moments — on the court and on the mat.
A few things I've built at the start.
Open to new opportunities and interesting problems. The fastest way to reach me is email.
mariia.monakhova.dev@gmail.com Email me