woladi

Hey, I’m Adrian 👋

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

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

💼 Available for consulting

Design Systems — component libraries, design tokens, Storybook, cross-team governance
AI-augmented DX — MCP servers, Figma Code Connect, streaming LLM apps in production
Privacy-first AI — local inference pipelines, GDPR-compliant LLM integrations

→ awolczuk@gmail.com · LinkedIn · GitHub

🏆 Selected Work

Enterprise PWA · CGI

Architected a modern insurance PWA achieving a 96% 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 Figma Code Connect to sync design tokens with production code

Production SaaS · Solo build

Full-stack subscription web app shipped to production. Stack highlights:

Streaming LLM responses — Vercel AI SDK + Claude, token-by-token rendering
Custom auth bridge — Clerk JWT decoded inside Supabase RLS policies (non-standard integration)
Heavy compute in serverless — C library compiled to WASM, singleton-cached across requests
Subscription billing — Stripe Checkout with dynamic pricing and promo codes

🛠 Open Source — Privacy-First AI Ecosystem

One philosophy across all repos: AI should work for you, not harvest your data. Local inference.

The privacy proxy

pseudonym-mcp — replaces PII with reversible tokens ([PERSON:1], [PESEL:1], [CREDIT_CARD:1]) before your prompt 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

The vision stack

Node.js Apple Vision Middleware four-layer ecosystem for local OCR & image analysis on macOS — same engine, different surfaces:

macos-vision — the foundation. Apple Vision OCR for Node.js, native and fast.
macos-vision-mcp — MCP server for Claude Code & Claude Desktop. Pre-extracts text before your AI sees it, cutting token usage by ~97% on real documents.
AiSort (early stage) — auto-tags and annotates files using the stack above. Files never leave your machine.

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

✍️ Writing

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.
Your Messaging Apps as a Private Document AI — Powered by OpenClaw — using macos-vision-mcp + pseudonym-mcp inside OpenClaw to OCR and anonymise documents sent over WhatsApp, Telegram, or Slack before they reach any cloud LLM.
Apple Vision vs Tesseract — A 50-File OCR-to-Markdown Benchmark — head-to-head OCR comparison on 50 PDFs with identical LLM formatter input; Tesseract wins on CER, Apple Vision wins on structural quality.
Privacy Tiers for Document AI — Three Pipeline Configurations — fully local vs local OCR + cloud LLM vs local OCR + pseudonymisation + cloud LLM: what each configuration protects, where the trade-offs are, and how to choose.
LLM as a Bridge Between Qualitative and Quantitative Research in Psychology and Sociology — using LLM-as-coder and LLM-as-judge to classify interviews, project free-form text onto validated scales (PHQ-9, BDI-II, SF-36), and correlate the result with survey data — with bias, non-determinism, and IRB caveats.