Front-end Tech Lead · React & Typescript · genAI/DX

10+ years building enterprise front-end systems at CGI. I specialize in closing the gap between design and engineering — through scalable component libraries, AI-augmented workflows, and developer experience that actually ships.

Open to consulting on design system architecture or AI-augmented DX pipelines. → Let's talk

• Design Systems → Scalable UI libraries with Storybook, design tokens, and cross-team governance
• Headless CMS → Storyblok architecture that empowers content teams without dev bottlenecks
• AI-Driven Engineering → MCP, Figma Code Connect, automated design-to-code pipelines

Architected a modern insurance PWA at CGI achieving a 95% Google Page Speed score:

• Component library with full Storybook documentation and testing
• Storyblok integration enabling visual editing independent of dev cycles
• AI-augmented delivery pipeline using MCP + Figma Code Connect to sync design tokens with production code

All projects share a common philosophy: AI should work for you, not harvest your data. Local inference. No cloud APIs. No subscriptions.

Local privacy proxy for LLMs — replaces PII with reversible tokens ([PERSON:1], [PESEL:1], [CREDIT_CARD:1]) before your prompt ever reaches the cloud, then restores original values in the response. GDPR-compliant pseudonymisation, fully offline, works with Claude, GPT-4, Gemini and any MCP-compatible client.

claude mcp add pseudonym-mcp -- npx pseudonym-mcp

MCP server for Claude Code & Claude Desktop — local OCR and image analysis via Apple Vision. Files never leave your Mac: no cloud API, no API keys, no network requests. Pre-extracts text before your AI ever sees it, cutting token usage by ~97% on real documents.

claude mcp add macos-vision-mcp -- npx macos-vision-mcp

Convert images and PDFs to structured Markdown using Apple Vision OCR + a local Ollama model — fully offline, no cloud APIs, no subscriptions.

npm install macos-vision-md

Apple Vision OCR & image analysis for Node.js — native, fast, offline, no API keys required.

npm install macos-vision

Automatically tags and annotates files using macos-vision-mcp and local LLMs — no cloud, no subscriptions. Files never leave your machine. Feed it a better model and it can also suggest smarter folder structures for your documents.

Your Obsidian Vault as a Private Second Brain — Powered by Local AI — how to use macos-vision-mcp + pseudonym-mcp to query your vault with cloud AI without exposing personal data.